implementation of an iec 61850 sampled values based line
TRANSCRIPT
Implementation of an IEC 61850 Sampled
Values Based Line Protection IED with a New
Transients-Based Hybrid Protection
Algorithm
By
Kapuge Kariyawasam Mudalige Sachintha Kariyawasam
A thesis submitted to the Faculty of Graduate Studies of
The University of Manitoba
in partial fulfilment of the requirements of the degree of
MASTER OF SCIENCE
Department of Electrical and Computer Engineering
University of Manitoba
Winnipeg, MB, Canada.
Copyright © April 2016 by Kapuge Kariyawasam Mudalige Sachintha
Kariyawasam
ii
Abstract
Over the course of the last decade, IEC 61850 has become the leading
standard for electrical substation automation the world over. This thesis
presents the work done towards implementing an IEC 61850 sampled values
based line protection Intelligent Electronics Device (IED) on a desktop
computer. This IED is capable of subscribing, decoding and processing the
sampled values published by a source to be used in line protection algorithms.
The novel hybrid line protection algorithm implemented in the IED
combines a distance protection scheme with a transients-based unit protection
method that uses sampled values complying with IEC 61850-9-2-LE.
Application of the line protection relay for protecting a series compensated
transmission line was examined using hardware-in-the-loop simulations.
Results indicate that (i) the implemented algorithms are robust and reliable,
and (ii) the class of transients based protection algorithms examined in this
thesis can be successfully implemented with IEC 61850 sampled values.
iii
Acknowledgments
During my stay at University of Manitoba over the last two years, I have
had the luxury of finding the support of many, without whom the successful
completion of this thesis would not have been possible. Therefore, it is only
decent to mention a few of them at least.
First and foremost, I owe my sincerest gratitude to my advisor, Dr. Athula
Rajapakse, who guided me all the way and supported me in every possible aspect
in completing this work. I’m also grateful to the Faculty of Graduate Studies,
University of Manitoba and Natural Sciences and Engineering Research Council
of Canada for supporting me financially throughout. Further, I would like to
convey my heartiest thanks to the examining committee for accepting this thesis
for review.
I would also like to acknowledge the financial, equipment, and technical
assistance provided by Pauley Family Foundation, USA, RTDS Technologies Inc.,
Canada, Schweitzer Engineering Laboratories, USA, ERL Phase Power
Technologies, Canada and Kalkitech Pvt. Ltd, India.
The ever-supportive academic and technical staffs of the Department of
Electrical and Computer Engineering should be specially mentioned for their
extended support. My loving family and caring friends must not go without a
mention, for they were the ones who made my life in Winnipeg enjoyable.
Finally, I would like to pay my tribute to each and everyone, who shared
their thoughts of wisdom and knowledge under difficult circumstances to make
this endeavor a success.
Sachintha Kariyawasam
iv
Dedication
To my loving parents.
v
Contents
Abstract .......................................................................................................... ii
Acknowledgments ......................................................................................... iii
Dedication ..................................................................................................... iv
Contents ......................................................................................................... v
List of Tables ................................................................................................. ix
List of Figures ................................................................................................ x
List of Abbreviations .................................................................................. xiv
List of Appendices ....................................................................................... xvi
1 Introduction 1
Background ............................................................................................ 1
1.1.1 Substation Automation ............................................................... 1
1.1.2 Substation Architecture of IEC 61850 ....................................... 3
1.1.3 GOOSE and SV ........................................................................... 4
1.1.4 Present Status of the Standard IEC 61850 ............................... 5
Motivation .............................................................................................. 6
Objectives of the Thesis ......................................................................... 7
Thesis Outline ........................................................................................ 8
2 Literature Review 9
Introduction to the Standard IEC 61850 .............................................. 9
2.1.1 Inception ...................................................................................... 9
2.1.2 Structure and Scope of IEC 61850 ........................................... 11
vi
2.1.3 Terms and Definitions .............................................................. 12
2.1.4 Key Advantages and Features of IEC 61850 ........................... 13
2.1.5 System Configuration description Language (SCL) ................ 16
2.1.6 Logical Interfaces ...................................................................... 18
2.1.7 Generic Object Oriented Substation Event ............................. 22
2.1.8 Sampled Values ......................................................................... 24
2.1.9 IEC 61850-9-2 LE ..................................................................... 25
2.1.10 Merging Unit ............................................................................. 26
2.1.11 Inter-Substation Communication as per IEC 61850-90-5 ....... 28
Transmission Line Protection Schemes .............................................. 29
2.2.1 Distance Protection ................................................................... 29
2.2.2 Challenges in Protecting Series Compensated Transmission
Lines 30
2.2.3 Current Transients Based Line Protection .............................. 32
Chapter Summary ............................................................................... 36
3 Implementation of a Process Bus Compatible IED Using Generic
Hardware 37
Experimental Setup ............................................................................. 37
Line Protection IED Application ......................................................... 40
3.2.1 Implementation of IEC 61850 Services ................................... 42
3.2.2 Digital Mimic Filter .................................................................. 43
3.2.3 Phasor Extraction using Discrete Fourier Transform ............ 46
IEC 61850 Communication Configurations ........................................ 50
3.3.1 Publication and Subscription of Information in IEC 61850 .... 50
3.3.2 Network Interface Card of the RTDS ...................................... 53
3.3.3 SV Publication from the RTDS ................................................. 55
3.3.4 SV Subscription by the Line Protection IED ........................... 57
3.3.5 GOOSE Publication from the Line Protection IED ................. 59
3.3.6 GOOSE Subscription by the RTDS .......................................... 59
vii
Time Delays in the Setup .................................................................... 61
Chapter Summary ............................................................................... 62
4 Distance Protection Scheme 63
Distance Estimation Units .................................................................. 63
Impedance Characteristics .................................................................. 65
4.2.1 Mho Characteristic ................................................................... 65
4.2.2 Quadrilateral Characteristics .................................................. 67
Fault Identification .............................................................................. 69
Validation of the Implemented Distance Protection Scheme ............ 70
Results Comparison for Impedance Estimations ............................... 72
Estimated Impedance Trajectories ..................................................... 74
Fault Detection Time ........................................................................... 76
Results for Series Compensated Transmission Lines ........................ 79
Chapter Summary ............................................................................... 81
5 Transients Based Protection Scheme 82
Concept of the Proposed Method of Line Protection .......................... 83
Implementation of the Transient Protection Scheme ........................ 85
5.2.1 Clarke Transformation of the Line Currents .......................... 85
5.2.2 Performing the Discrete Wavelet Transform ........................... 86
5.2.3 Fault Identification ................................................................... 92
Hybrid Protection Algorithm .............................................................. 93
Validation of the Hybrid Protection Scheme ...................................... 97
Performance Evaluation of the Hybrid Protection Scheme ............. 110
5.5.1 Behaviour for Non-faulty Conditions ..................................... 111
Effect of the Bit Resolution of the A/D Converter ............................ 113
Chapter Summary ............................................................................. 117
6 Conclusions 118
viii
Contributions and Conclusions ......................................................... 118
Future Work ...................................................................................... 122
7 References 124
8 Appendix A 131
9 Appendix B 147
10 Appendix C 150
11 Appendix D 151
ix
List of Tables
Table 1: Comparison of phase unit impedance calculations ........................... 72
Table 2: Comparison of ground unit impedance calculations ......................... 73
Table 3: Fault detection time in terms of number of samples for distance
protection algorithm ............................................................................ 78
Table 4: Performance of the proposed protection scheme under different fault
conditions ........................................................................................... 110
x
List of Figures
Figure 1: Typical substation architecture of IEC 61850 ................................... 3
Figure 2: Flow of information in the configuration process ............................ 17
Figure 3: Logical interfaces for substation communication in IEC 61850 [6] 19
Figure 4: Typical substation topology of IEC 61850 ........................................ 20
Figure 5: A representation of IEC 61850 sampled values protocol ................. 24
Figure 6: Function of a merging unit [15] ........................................................ 27
Figure 7: Impedance seen by a distance protection relay ............................... 29
Figure 8: Distance characteristics on R-X (resistance-reactance) plane for three
zones; (a) Mho, (b) Quadrilateral (line impedance is plotted in dashed line) . 30
Figure 9: Voltage and current inversion in a series compensated transmission
line ..................................................................................................................... 31
Figure 10: Wavelet function of (a) ‘db4’ mother Wavelet, (b) ‘Symlets 2’ mother
wavelet ............................................................................................................... 34
Figure 11: Multi-level filter bank implementation of DWT; (2↓) denotes down-
sampling by 2 .................................................................................................... 35
Figure 12: Schematic of the apparatus for the distance protection scheme ... 38
Figure 13: Implementation structure of the line protection IED; only the
distance protection function is illustrated ....................................................... 41
Figure 14: (a) Unfiltered current waveform, (b) Filtered current waveform . 45
Figure 15: Magnitude and phase of the extracted current phasor plotted
against the sample number .............................................................................. 49
Figure 16: Physical connections in the LAN .................................................... 54
xi
Figure 17: Anti-aliasing filters (left) and the sampled values component (right),
residing in the RTDS simulation case .............................................................. 56
Figure 18: Signal mapping for SV subscription ............................................... 58
Figure 19: An example signal mappings for GOOSE subscription in; (a)
Distance Protection scheme (b) Transients based protection scheme ............ 60
Figure 20: Illustration of the implementation of a breaker IED in the RTDS
simulation .......................................................................................................... 61
Figure 21: Implementation of zone 1 Mho characteristic ................................ 65
Figure 22: Implementation of zone 1 quadrilateral characteristic ................. 67
Figure 23: Fault Identification algorithm of the distance protection scheme 69
Figure 24: SLD of the simulated power system using which the implemented
distance protection scheme is validated ........................................................... 70
Figure 25: Schematic of the setup for testing the distance protection scheme
........................................................................................................................... 71
Figure 26: Trajectory of the estimated impedance of the line protection IED for
a zone 1, AB fault, plotted on a R-X plane; Zone 1 Mho characteristics and
impedance line are shown................................................................................. 75
Figure 27: Fault detection in the distance protection scheme for a 10 % fault
........................................................................................................................... 76
Figure 28: Fault detection in the distance protection scheme for a 90 % fault
........................................................................................................................... 77
Figure 29: Estimated impedance trajectories for a bus 1 fault (a) with, and (b)
without series compensation. Zone 1 Mho characteristic is also shown. ........ 80
Figure 30: Estimated impedance trajectories for a bus 1 fault (a) with, and (b)
without series compensation. Zones 1 and 2 Mho characteristic are also shown.
........................................................................................................................... 81
Figure 31: Origination of high frequency wave-fronts due to a fault ............. 83
Figure 32: Implementation of the DWT ........................................................... 87
Figure 33: Extracted high frequency transient from Iα (for a sampling
frequency 80 samples per cycle) ....................................................................... 87
xii
Figure 34: Extracted high frequency transient from Iα (for a sampling
frequency 256 samples per cycle) ..................................................................... 88
Figure 35: Detail wavelet coefficients at the near end for the same fault with a
sampling rate of (a) 256 samples per cycle (b) 80 samples per cycle .............. 89
Figure 36: Detail wavelet coefficients at near end for (a) a 10% fault with fault
inception angle at 0°, (b) a 10% fault with fault inception angle at 180°, (c) a
50% fault with fault inception angle at 0° and (d) a 90% fault with fault
inception angle at 0°. ......................................................................................... 90
Figure 37: Detail wavelet coefficients at either end for a 50% fault ............... 91
Figure 38: Absolute values of the detail wavelet coefficients at either end for a
50% fault ............................................................................................................ 91
Figure 39: Sending of GOOSE messages corresponding to wave-front polarity
........................................................................................................................... 93
Figure 40: Illustration of the transfer-trip scheme ......................................... 94
Figure 41: Modified Mho characteristics for the proposed protection method95
Figure 42: Combined algorithm of fault detection for the proposed hybrid
protection scheme .............................................................................................. 96
Figure 43: SLD of the simulated power system using which the implemented
transient protection scheme is validated ......................................................... 97
Figure 44: Schematic of the developed system for the hybrid protection scheme
........................................................................................................................... 98
Figure 45: Three phase currents, voltages, model current component (Iα), detail
WTCs, trajectories of the estimated impedance and the fault detection at either
end for a 50%, ABC fault. ............................................................................... 100
Figure 46: Voltage measurements at the bus side and the line side ............ 101
Figure 47: Trajectories of the estimated impedance and detail wavelet
coefficients at either end for a bus 1, ABC fault, with the effect of voltage
inversion .......................................................................................................... 102
Figure 48: Trajectory of the estimated impedance for a 5% three phase fault
with series capacitor in service, (with the effect of voltage inversion) ......... 103
xiii
Figure 49: Detail wavelet coefficients at either end for a 5% three phase fault
with series capacitor in service, (with the effect of voltage inversion) ......... 103
Figure 50: Detail wavelet coefficients at either end for a 10% AB fault ...... 104
Figure 51: Detail wavelet coefficients at either end for a 50% AB fault ...... 105
Figure 52: Detail wavelet coefficients at either end for a 90% AB fault ...... 105
Figure 53: Detail wavelet coefficients at either end for an external fault (a bus
1 fault) ............................................................................................................. 106
Figure 54: Detail wavelet coefficients at either end for a 5% AB fault with 40%
series compensation ........................................................................................ 108
Figure 55: Detail wavelet coefficients at either end for a 95% AG fault with a
50 Ω fault impedance ...................................................................................... 108
Figure 56: Fault detection in the proposed protection scheme for a solid AB
fault, 5% away from bus-1. ............................................................................. 109
Figure 57: Detail wavelet coefficients at either end for a closing operation of
the circuit breaker at the near end ................................................................ 111
Figure 58: Detail wavelet coefficients at either end for a closing operation of
the bypass switch of the series capacitor. ...................................................... 112
Figure 59: Detail wavelet coefficients at the near end for a 50% fault, with an
assumed A/D converter bit resolution of (a) 24 bits, (b) 18 bits (c) 16 bits and
(d) 12 bits ......................................................................................................... 115
Figure 60: Modelling the effect of a 24 bit A to D converter ......................... 116
xiv
List of Abbreviations
BIED Breaker Intelligent Electronic Device
CID Configured IED Description
CT Current Transformer
DB4 Daubechies 4
DFT Discrete Fourier Transform
DNP3 Distributed Network Protocol
DWT Discrete Wavelet Transform
EMT Electromagnetic Transient
FCDA Functionally Constrained Data Attributes
Gbps Gigabits per Second
GPS Global Positioning System
GOOSE Generic Object Oriented Substation Event
HMI Human Machine Interface
ICD IED Capability Description
IEC International Electro-Technical Commission
IED Intelligent Electronic Device
IID Instantiated IED Description
LAN Local Area Network
xv
LN Logical Node
MAC Media Access Control
Mbps Megabits per Second
MU Merging Unit
NIC Network Interface Card
PMU Phasor Measurement Unit
PPS Pulse per Second
RTDS Real Time Digital Simulator
SA Substation Automation
SAS Substation Automation System
SCD Substation Configuration Description
SCL System Configuration description Language
SCSM Specific Communication Service Mapping
SED System Exchange Description
SLD Single Line Diagram
SSD System Specification Description
SV Sampled Values
VLAN Virtual Local Area Network
VT Voltage Transformer
WAN Wide Area Network
XML Extensible Markup Language
WT Wavelet Transform
xvi
List of Appendices
Appendix A:
I. IEC 61850 Service Modules
II. CID file of the RTDSTM for SV publication
III. CID file of the Line Protection IED
Appendix B:
I. Transmission Line Parameters
II. Series Capacitor Parameters
Appendix C: Filter coefficients for the mother wavelet “Daubechies 4 (db4)”
Appendix D:
I. Behaviour of traveling waves at a junction on a
transmission line
II. Analysis of Wave-front polarities for a R-L circuit
Chapter 1
Introduction
Background
IEC 61850 is a relatively new standard for electrical substation
automation. It was introduced in early 2000s by the International Electro-
technical Commission responding to the long existed necessity to standardize
substation automation practices. In recent years, many power utilities around
the world have implemented IEC 61850, often partially or on trial basis, in
their grids. However, IEC 61850 is fast becoming the leading standard for
substation automation [1].
1.1.1 Substation Automation
Substation Automation (SA), on the subject of electrical engineering,
refers to the automatic real-time operation and control of a power system. Main
tasks of substation automation include data acquisition, power system
monitoring, power system control and communication among, intelligent
2
electronic devices (IED). These SA functions should be implemented in such a
way that the necessary operational efficiency is maintained with minimal
human intervention and engineering resources. Substation automation has
been an integral part of power system operation for decades. Among the key
components of an automated electrical substation are meters, sensors,
actuators, control equipment, human-machine interface equipment, and the
communication network [2], [3].
The communication network plays a crucial role in a substation
automation system, and the communication protocols enable various devices
connected to the communication network to properly decode the information
received [4]. Over the years, vendor specific proprietary protocols as well as a
number of standard protocols such as Modbus, DNP3 and IEC 60870-5-104
have been utilized for implementation of substation automation systems.
Numerous flaws in these standards, such as the lack of interoperability among
equipment from different vendors, prompted the inception of the standard IEC
61850 [1]. It unifies all substation automation practices into one single
standard [4], [5]. IEC 61850 provides explicit procedures for design, modelling,
data representation, and IED configurations, covering all aspects of a SA
system. As a result, the degree of interoperability among IEC 61850 compliant
IEDs is far superior to that of their non-compliant counterparts, a fact that is
often cited as a major advantage of the standard [4].
3
1.1.2 Substation Architecture of IEC 61850
IEC 61850 introduces a distinctive substation architecture for
communication within the SA system. In its most commonly encountered form,
the IEC 61850 substation architecture contains three levels and two
communication bus systems as shown in Figure 1. The three levels are
referred to as process level, bay level and station level, whereas the two buses
are referred to as process bus and station bus [6].
Figure 1: Typical substation architecture of IEC 61850
Hardwired connections
HMI Station
Computer Gateway
Control
IED
Protection
IED
Merging
Unit
Breaker
IED
Control
IED
Protection
IED
Breaker
IED
Merging
Unit
Switchgear (primary plant)
Station
Level
Station
Bus
Bay
Level
Process
Bus
Process
Level
Switch
Switch
4
The process level comprises of devices, which interface with the primary
plant (switchgear and instrument transformers) such as merging units (MU),
breaker IEDs (BIED), sensors, and actuators.
The bay level comprises of all the protection, control and monitoring
IEDs. And the station level typically consists of a Human Machine Interface
(HMI), station computers, a database and remote communication interfaces.
The station bus enables communication between the station level and
the bay level, along with inter-IED communication within the station/bay
levels. The process bus facilitates the communication between the process level
and the bay level. This includes the exchange of instantaneous raw data (such
as current and voltage transformer measurements) as well as control
information between process and bay levels [6].
1.1.3 GOOSE and SV
GOOSE (Generic Object Oriented Substation Event) and SV (Sampled
Values) are two of the main specific communication services mappings (SCSM)
defined in IEC 61850, in parts 8-1 and 9-2, respectively [7], [8]. Generally,
GOOSE messages are employed for exchange of fast control information
between IEDs. Moreover, GOOSE messages are time critical, uni-directional,
multicast messages with no acknowledgements. Even though GOOSE
messages are primarily used to indicate changes in the state of substation
parameters, they can carry both binary and analog values.
5
SV, in contrast, is the protocol for acquisition of raw information
particularly that is measured by instrument transformers. SV streams carry
digitized samples of the analog measurements at a defined sampling rate.
These messages are also time critical. The special IED that converts the analog
readings from the instrument transformers into SV packets is called a Merging
Unit (MU). A majority of vendors follow the UCA International Users Group
Implementation Guidelines, better known as IEC 61850-9-2 LE, for the
implementation of SV protocol. According to 9-2 LE, a SV packet carries 8
instantaneous values; three phase and neutral values for both voltage and
current at a specified sampling rate of either 80 or 256 samples per cycle.
Further details on the IEC 61850 substation architecture and the sampled
values will be provided in subsequent chapters of the thesis.
1.1.4 Present Status of the Standard IEC 61850
The standard IEC 61850 comprises of ten parts. There are two editions
of the standard; edition 2 being is the latest. With the integration of modern
communication technologies into substation automation systems,
implementation of IEC 61850 is becoming an inevitable eventuality; many
utilities around the world have already shifted towards IEC 61850 [1]. It is
widely accepted and implemented in Europe and Asia. Implementation of IEC
61850 is making a relatively slow progress in North America.
6
Motivation
Majority of the IEC 61850 implementations in real-world operational
SAS system are restricted to the station bus. Many utilities around the world
have incorporated IEC 61850 station bus communication into their girds; at
least to a certain extent. Application of GOOSE messages in SA functions
including event recording, status updating and interlocking is widespread,
while its use in more critical tasks such as breaker tripping is less common.
On the other hand, many of the SV implementations still remain at
testing and research levels, even-though SV based digital process bus
communication is an integral part of IEC 61850. This is due to many reasons
including lack of SV compatible IEDs (at both process and bay levels),
unreliability caused by extensive network traffic and high bandwidth demand.
Although many of the commercially available protection IEDs are IEC
61850 station bus compliant, commercial protection IEDs that use SV (process
bus compatible IEDs) are not very common. In addition, many aspects of SV
communication remains unexplored and much of its potential is unexploited.
SV presents an opportunity of innovation of protection philosophies: for
example it allows development of distributed applications based on current and
voltage values communicated between devices connected to the process bus.
Also, the high sampling rates associated with IEC 61850-9-2 LE
implementation allows development of protection algorithms based on high
7
frequency transient signals. These prospects are the factors that motivated
this research.
Objectives of the Thesis
An SV compatible IED porotype developed in-house, not only presents
the opportunity to study various practical aspects of SV communication but
also allows exploring the potential for implementing innovative protection
algorithms. A test setup fully compatible with IEC 61850, including the digital
process bus, is needed for testing SV compatible protection IEDs and SV
communication. Development of such a test setup has additional long lasting
usages for future research, teaching and training. Thus, the primary objectives
of this thesis are:
1. Developing a line protection IED that is compatible with Sampled
Values protocol as per IEC 61850-9-2. This IED implements the basic
functionality of a distance protection relay.
2. Incorporating a transient based protection algorithm that is not
normally used in traditional line protection relays into the developed
IED to improve the reliability when protecting series compensated
transmission lines.
3. Integrate the developed SV compatible prototype IED to a test setup
developed in [9] to complete a fully IEC 61850 compliant laboratory.
8
Thesis Outline
The subsequent sections of this thesis are organized as follows. Chapter
2 presents a comprehensive introduction to the IEC 61850 standard,
emphasizing on SV protocol and process bus communication. Further, brief
reviews on distance protection and current transients based line protection are
also provided. In Chapter 3, development of the line protection IED is
extensively explained, providing details on its structure, internal algorithms
and communication configurations.
In Chapters 4 and 5, implementations of the two protection schemes;
distance protection and transient based line protection schemes, are presented,
respectively, with corresponding results for validation. The protection
algorithms are tested for a variety of faults and for varied system parameters.
Moreover, the power system simulated in a Real Time Digital Simulator (on
which the developed IED is tested), is also described in this chapter. Finally,
Chapter 6 presents the conclusions and the future work of the research.
Chapter 2
Literature Review
Introduction to the Standard IEC 61850
2.1.1 Inception
Communication between IEDs is of paramount importance to substation
automation and therefore, to the overall operation and control of the power
system. Few of the fundamental requirements of a communication system of
an electrical substation include,
Fast inter-IED communication
Multi-vendor interoperability
High availability
Sufficient security
Assured delivery times
10
With the internet and other communication technologies in rapid progression,
the need for standardization of substation automation practices became
essential [4].
Prior to IEC 61850, SA systems were often developed using vendor-
specific protocols, which essentially lacked interoperability as well as were
inherently inflexible and limited. Development of SA systems based on these
protocols consumed countless hours of tedious work for designing, testing and
troubleshooting. Therefore, the need for standardized, interoperable means of
substation automation was ever present among the electrical automation
engineers the world over [1], [3]. The standard IEC 61850 came into existence
in early 2000s, owing to the work done in the IEC (International Electro-
technical Commission) Technical Committee Number 57 (TC57), Working
Group 10 (WG10) [4].
The initial objectives, which the standard was intended to serve
included, unification of all substation automation disciplines into one
standard, enhancement of multi-vendor interoperability, standardization of
data representation and introduction of necessary procedures for equipment
testing [4], [5].
11
2.1.2 Structure and Scope of IEC 61850
IEC 61850 standard comprises 10 main parts, which are as listed below [5],
IEC 61850-1: Introduction and overview
IEC 61850-2: Glossary
IEC 61850-3: General requirements
IEC 61850-4: System and project management
IEC 61850-5: Communication requirements for functions and device
models
IEC 61850-6: Configuration language for communication in electrical
substations related to IEDs
IEC 61850-7: Basic communication structure for substation and feeder
equipment
IEC 61850-7-1: Principles and models
IEC 61850-7-2: Abstract communication service interface (ACSI)
IEC 61850-7-3: Common Data Classes
IEC 61850-7-4: Compatible logical node classes and data classes
IEC 61850-7-10: Communication networks and systems in power
utility automation - Requirements for web-based
and structured access to the IEC 61850 information
models [Approved new work]
IEC 61850-8: Specific communication service mapping (SCSM)
IEC 61850-8-1: Mappings to MMS (ISO/IEC9506-1 and ISO/IEC
9506-2)
IEC 61850-9: Specific communication service mapping (SCSM)
IEC 61850-9-1: Sampled values over serial unidirectional multi-
drop point to point link
IEC 61850-9-2: Sampled values over ISO/IEC 8802-3
IEC 61850-10: Conformance testing
12
2.1.3 Terms and Definitions
IEC 61850 provides many terms and definitions related to its scope.
Following are few of the most important among them, which are used
frequently in the following sub-sections of this thesis [6].
2.1.3.1 Primary System
Primary system or primary plant refers to all the power system equipment and
switchgear contained in an electrical substation [6].
2.1.3.2 Substation Automation System (SAS)
Substation automation system operates, controls, protects and monitors the
primary system using the substation communication system [6].
2.1.3.3 Function
With reference to IEC 61850, a function is simply a task executed by the SA
systems. These functions oversee the control, protection, monitoring and
maintenance of the SA systems. Logical nodes exist as subsets of functions
(this is further explained under logical nodes). Some functions are performed
by more than one logical node, which might be placed in different IEDs. These
functions are referred to as distributed functions [6].
2.1.3.4 Logical Node (LN)
A logical node is the smallest part in a function that can exchange data. By
definition, only LNs can exchange data (consequently, the only data that can
13
be exchanged is that resides within a LN). A LN of a power system equipment
represents the intelligent body of that device in the SA system (such as a circuit
breaker) and it contains its own data. Additionally, within an IED, it
represents and implements a function (such as a distance protection scheme).
Communication links between LNs are called logical connections [6].
2.1.3.5 Intelligent Electronic Device (IED)
An intelligent electronic device or an IED is a device that possesses one or more
electronic processors that are capable of receiving/sending information from/to
an external device. It can implement the functionalities of the logical nodes
above specified in a specific context. IEDs contain internal clocks to provide
necessary time tags. Protection and control (digital) relays are classic examples
of IEDs. The term “physical device” has the same meaning as “intelligent
electronic device” in the context of IEC 61850 [6].
2.1.4 Key Advantages and Features of IEC 61850
IEC 61850 has several unique features and advantages that set it apart
from legacy SA protocols. Few of the most important among them are briefly
explained below.
2.1.4.1 Enhancement of Multi-Vendor Interoperability
The word interoperability, in the context of IEC 61850, refers to the
capability of IEDs of the same or different manufacturers to exchange
14
information among each other and to apply that information to acquire the
desired coordination. Manufactures can achieve this by adhering to the explicit
definitions, specifications and procedures provided in the standard.
Substation automation protocols prior to IEC 61850 have had several
issues regarding interoperability. Manufactures sometimes have their own
proprietary protocols for substation automation functions. This makes the
configurations between devices of different vendors a tedious exercise. Usually,
utilities either have to stick with one vendor or use protocol convertors [1]. In
contrast, all IEC 61850 compliant devices, once configured, are capable of
working in tandem with each other regardless of their make. This essentially
simplifies the engineering tasks involved, and also enables the selection of
devices based on performance rather than compatibility.
2.1.4.2 Simplification of Hardwired Implementations
The next vital advantage IEC 61850 offers is the simplification of
hardwired engineering implementations. Older SASs require a huge number
of physical connections, both for communication between devices and for raw
data acquisition. Significant amounts of time and money must be consumed in
designing, configuring and troubleshooting these connections. With IEC 61850,
the number of connections and associated configurations are considerably
reduced, as all communication is done digitally. As a result, IEC 61850 based
SASs are more or less copper-free, which reduces the overall costs [10].
15
Moreover, addition or modification of equipment in legacy substations requires
physical reconfiguration of the system, whereas in IEC 61850 SASs, all the
required new configurations can be performed on software and sent to IEDs
via the communication network.
2.1.4.3 Organized Substation Modelling and Data Representation
Object oriented data representation and object modelling IEC 61850
offers, makes it far more intuitive and user friendly than its counterparts. Data
representation (data models) in IEC 61850 emulates the structure (topology)
of the substation it’s intended for [4], [11]. This concept is referred to in the
standard as virtualization. Data representation in other substation
automation protocols do not specifically represent the physical substation
entity that it represents. This again makes design and configuration with IEC
61850 easier, which in a SA system with hundreds of IEDs could make a
substantial difference.
IEC 61850 put forth meticulous engineering procedures, naming
conventions and equipment models for the design and configuration of an
entire substation. IEC 61850 follows a unified approach in developing the SAS.
This starts from the specification of requirements and gradually evolves into a
descriptive virtual model of the actual substation. Another unique feature of
IEC 61850 is its relative independence from the prevailing communication
technology. In IEC 61850, the basic modelling parts are decoupled from the
16
methods of information exchange (i.e. it uses abstract modelling). In that way,
the core of the standard will remain intact, while its implementations can
adopt the state-of-the-art communication technologies [4].
2.1.5 System Configuration description Language (SCL)
The XML based Substation Configuration description Language (SCL)
is introduced in IEC 61850-6, for the purpose of formal documentation of the
configuration flow in a SAS. This is used for the description of configurations
of IEDs, communication systems as well as for defining the connection between
the SAS and the physical substation itself [12]. It further facilitates the
exchange of engineering configurations back and forth between different
stages of the design process of the SAS in correct coordination, thus, ensuring
interoperability.
In part 6 of the standard, two conceptual tools are introduced for the
purpose of configuring the SAS, which are, IED configurator and system
configurator. IED configurators are mostly vendor-specific. They are capable
of generating IED template files, setting-up IED specific configurations and
loading files into the IEDs. In contrast, a system configurator is independent
of the IEDs (and the vendors) and used for higher level system engineering. It
imports configuration files from different IEDs and creates a substation-
related configuration file with shared system information. This file is then
taken into the IED configuration tools for final IED related configurations. For
17
the purpose of abovementioned configuration information exchange, SCL
defines six file types with the extensions ICD, IID, SSD, SCD, CID, and SED
[12]. The basic flow of information in the configuration process is depicted in
Figure 2 [12].
Figure 2: Flow of information in the configuration process
Two types of SCL files mentioned above are used exclusively for setting
up the IEC 61850 configurations in IEDs. They are the ICD (IED Capability
Description) files and the CID (Configured IED Description) file. An ICD file
can be considered as a template file of a certain IED, describing the
functionality (related to IEC 61850) of that IED. It contains data type
definitions, logical nodes type definitions and sometimes substation templates
as well. All commercially available IEDs come with an ICD file provided by the
manufacturer. A CID file, on the other hand, is describing the IEC 61850
SCD
file CID
file
IED Configuration
Tool
System
Configuration
Tool
SSD
file
ICD
file
IED
IID
file
System
Configuration
Tool
SED
file
CID
file
IED
18
communication configurations of an instantiated IED. It is created by the IED
configurator from a larger, overall substation based file called SCD (System
Configuration Description) file.
The object model in IEC 61850 consists of three parts, namely,
substation model, product model and Communication model. The substation
model is a hierarchy of virtual objects (data models) reproducing the physical
substation. It defines the topology of switchyard equipment and their
functional designations. The product model (or the IED model) is also a
hierarchy of data models, which represent the functional aspects of the IEDs
that form the SAS. The communication model is not a hierarchy and it
describes the possible communication connections between IEDs. It comprises
of communication-specific data objects [12].
2.1.6 Logical Interfaces
Different logical interfaces in a SAS as defined in IEC 61850-5, are
illustrated in Figure 3 [6]. The corresponding logical structure of the SAS is
reproduced in Figure 4 for convenience. The interfaces (IFs) illustrated in
Figure 3 (as numbers) are [6],
IF 1: Communication between bay level and station level to exchange
protection data
IF 2: Communication between bay level and remote protection schemes
IF 3: Communication within bay level
19
IF 4: Communication between process and bay level to exchange CT and VT
instantaneous data (especially samples)
IF 5: Communication between process and bay level to exchange control data
IF 6: Communication between bay and station level to exchange control data
IF 7: Communication between substation (level) and a remote engineer’s
workplace
IF 8: Communication between bays (for fast functions such as interlocking)
IF 9: Communication within station level
IF 10: Communication between substation and a remote control centre to
exchange control data
Figure 3: Logical interfaces for substation communication in IEC 61850 [6]
FCT.A FCT.B
Remote Process Interface Sensors Actuators
Control Protection Protection Control
4, 5 4, 5
8
1, 6 1, 6
2
3 3
9
7 10
2
Technical Services
Process Level
Bay Level
Station Level
Remote Control
Remote Protection
Primary Plant
20
As shown in Figure 4, equipment that interface with the primary plant
(switchgear) such as merging units (MU), breaker IEDs (BIED), sensors and
actuators are contained in the process level. The bay level contains all
protection, control and monitoring IEDs, while the station level usually
consists of a Human Machine Interface (HMI), a station computer, a database
etc.
Figure 4: Typical substation topology of IEC 61850
The structure shown in Figure 3 refers to the logical structure of the
SAS given in Figure 4. The physical devices, however, may be installed in a
different structure. There is no definite connection between this logical
architecture and the physical arrangement of devices. These logical interfaces
can be implemented as separate physical interfaces or a couple of logical
interfaces can be integrated into one physical interface. For instance, some
process and bay level functions might be found inside the same physical device.
Hardwired connections
HMI Station
Computer Gateway
Control IED Protection IED
MU BIED
Control IED Protection IED
BIED MU
Switchgear (primary plant)
Station
Level
Station Bus
Bay
Level
Process
Bus
Process
Level
Switch
Switch
21
Moreover, one or more of these physical interfaces can be combined into a
single LAN. This physical architecture depends on factors like availability of
devices, performance requirements, cost, geography and other constraints. It
is also important to understand that not all of the logical interfaces are
compulsory in a SAS [6].
Using the logical interfaces given in Figure 3, two important LANs (or
bus systems) can be identified. According to IEC 61850-5, the combination of
the interfaces 1, 6, 3, 9 and 8 can be identified as the Station Bus, whereas
interfaces 4 and 5, together can be identified as the Process Bus [6].
2.1.6.1 Station Bus
The station bus facilitates communication between the station level and
the bay level as well as inter-IED communication within the station/bay levels.
It is usually used for SA functions such as interlocking, reporting and event
recording.
2.1.6.2 Process Bus
In an electrical substation, high voltage equipment and switchgear such
as power transformers, instrument transformers, busbars and circuit breakers
are referred to as the “primary plant”, while the protection, control and
monitoring equipment including relays, event recorders and other IEDs are
called the substation automation system (SAS). The process bus facilitates the
communication between IEDs that govern the primary plant and the main
22
protection and control devices of the SAS. The links in the process bus in a
traditional SAS usually are dedicated, function specific and hardwired copper
connections, connecting the primary plant to the IEDs at bay level.
A digital process bus as proposed in IEC 61850, however, only comprises
a single Ethernet network that carries digitized information of the measured
analog signals. It may also carry command signals, alarms and indicator
information like status updates from the relays to the primary plant as well. A
process bus is typically confined to a single bay only [6]. Occasions in which
process bus extends to more than one bay, it might govern the data exchange
between bay level IEDs as well.
2.1.7 Generic Object Oriented Substation Event
Generic Object Oriented Substation Event (GOOSE) is one of the two
main specific communication services mappings (SCSM) used in IEC 61850.
Defined in IEC 61850-8-1 [7], GOOSE is a method of exchanging messages
between IEDs over a local Ethernet network. It is both fast and reliable and
therefore, well suited for time-critical applications such as protection functions
in a substation.
Since GOOSE messages are involved with time-critical functions,
acknowledgements are not sent. Instead, to increase redundancy, the same
GOOSE message is retransmitted in a predefined pattern. Moreover, GOOSE
is a type of multicast messaging that can carry both binary and analog values
23
in their messages. However, in practice, GOOSE messages mainly carry
changes of state of substation parameters. GOOSE messages are suitable for
the exchange data involving SA functions such as breaker tripping, status
update, interlocking and event recording.
GOOSE messages are based on a publisher/subscriber model and since
there are usually multiple receivers involved, multicasting is required (i.e.
multicasting refers to sending information to a selected group of destinations,
simultaneously). Since, GOOSE are time critical messages, application layer
is directly mapped to data link layer, skipping the intermediate layers
including transport and network layers. Due to this, GOOSE messages involve
no IP, no addressing, and can be transmitted via a LAN only (are not routable
through a WAN).
From the above discussion, the following can be viewed as features of
IEC 61850 GOOSE messaging [13],
Uni-directional communication
Multicast messaging
Subscriber receives only the required (mapped) information; the rest
is neglected
Very fast communication
No acknowledgements sent upon reception of the message
Can be used only for intra-substation communication
24
2.1.8 Sampled Values
Sampled Values (SV) or Sampled Measured Values (SMV) is a protocol
defined in IEC 61850-9-2 for the acquisition of raw data [8]. In particular, it
facilitates the transfer of digitized samples of analog measurements. Similar
to GOOSE, SV are also time critical messages and can be streamed as either
unicast or multicast. The following can be viewed as the key features of SV,
SV are time critical messages, hence no acknowledgements are sent.
And as in GOOSE, link and application layers are directly mapped,
improving the time performance of data transfer. However, unlike in
GOOSE, the same message is not retransmitted in SV.
SV protocol continuously publishes data packets at a specific rate(s)
defined by the user (as depicted in Figure 5).
Figure 5: A representation of IEC 61850 sampled values protocol
A
B
C
D
SV Control Block
SV dataset
Time Synchronized data
SV packets continuously published
to the process bus at a specified rate
SV Control block
Parameters
SV Ethernet
packet
format
T0 T0+t T0+2t
Defined in the CID file
of the SV publisher
25
2.1.9 IEC 61850-9-2 LE
The UCAIug implementation guideline, often referred to as “9-2 Light
Edition” (9-2 LE) [14], specifies simplified guidelines for implementing IEC
61850-9-2 based process bus communication using SVs. With 9-2 LE, dataset
size, sampling rate, time synchronization requirements and the physical
interfaces to be used are predefined. An IEC 61850-9-2 LE compliant device
would publish eight simultaneous, digitalized data streams, four voltages and
four currents (three phases and neutral for each), at a sampling rate of 80 or
256 samples per cycle.
The function of an instrument transformer (i.e. measuring and
publishing of a particular parameter), is represented by a corresponding logical
node inside the ICD file of merging unit. For instance, the current transformer
attached to phase A is represented by an instance of current transformer
logical node class “TCTR”. Similarly, the voltage transformer function is
represented by voltage transformer logical node class “TVTR”. Therefore, an
ICD file of a MU compliant with 9-2 LE, comprises of four logical node
instances each, of classes TCTR and TVTR, in order to accommodate eight SV
streams published.
Further, MUs in SAS can be synchronized to an external time reference
(often to a GPS clock) and the synchronization signal must be a 1PPS input,
which shall have an accuracy of ±1 μs, according to the specification in IEC
60044-8. According to the standard, there can be a maximum propagation
26
delay of ±2 μs between the synchronization signal output and the MU clock
input, which otherwise ought to be compensated by the MU [14].
Moreover, if the neutral current/ voltage are not measured by the MU,
they should be calculated using phase measurements inside the MU. Also, this
should be informed to the receiver using the respective quality field. The
attribute “SmpSynch” indicates the state of the synchronization signal of the
MU. The attribute “SmpCnt” designates the sample number taken. It is
incremented at every new sample and is reset to zero with every synchronizing
pulse (i.e. at the start of each second) [14]. Therefore, for 60 Hz systems, this
variable moves from 0 to 4799 or 15359 depending on the sampling rate used.
This sample count can be used for the time synchronizing purposes in IEDs.
2.1.10 Merging Unit
A Merging unit (MU) is an IED that digitizes the analog measurements,
taken mainly by conventional current and voltage transformers. It then
publishes those data as a stream into the process bus at a predefined rate, to
be captured by protection and control relays at the bay level. MUs merge (or
combine) and perform time correlation of voltages and currents of the three
phases (of a line) as shown in Figure 6 and hence, given its name [15]. These
devices operate as per part 9 of the standard IEC 61850. Further, it is
important to notice that the connections from the instrument transformers to
merging units are usually hardwired.
27
Figure 6: Function of a merging unit [15]
Hardwired connections
Process bus (Ethernet network)
MERGING UNIT (MU)
Phase A Phase B Phase C
Time synchronization signal
Instrument
transformers
Sampled and merged analog signals
Published SV stream
Correlation in time
Time
28
2.1.11 Inter-Substation Communication as per IEC 61850-90-5
All communication mechanisms discussed thus far in this thesis with
reference to IEC 61850 are confined with in a particular substation (In other
words, IEC 61850 is basically for intra-substation communication). However,
a newly introduced part of the standard, IEC/TR 61850-90-5, stretches its
boundaries in an attempt to incorporate information exchange among
substations over a WAN as well [16].
This new part of the standard basically focuses on the transmission of
synchrophasor information according to IEEE C37.118, using IEC 61850. In
addition, this enables regular IEC 61850-8-1 GOOSE and IEC 61850-9-2 SV
packets to be routed over a WAN [17]. This new routable GOOSE messaging is
envisioned to be used in order to accomplish the transfer trip scheme proposed
in a later stage of this thesis.
29
Transmission Line Protection Schemes
2.2.1 Distance Protection
Distance protection is one of the most established practices in power
system protection. It is a non-unit type protection method and widely used for
transmission line protection. A distance protection relay estimates the
impedance down the line to a fault using currents and voltage measurements
at the relay point [18].
As illustrated in Figure 7, the impedance seen by the distance protection
relay at point A for a fault occurring at point B would be ZAB, irrespective of
the magnitudes of voltage and current during the fault. Owing to this, distance
protection has the capability to discriminate faults in different parts of the
power system. It is also fast operating and simple to implement [18].
Figure 7: Impedance seen by a distance protection relay
In general, a distance protection relay has two types of settings; reach
settings and operating time settings. The reach settings are applied as zones,
ZAB
Bus 2 Bus 1
F1
B A
Distance
Protection Relay
CT VT
30
usually 3 or 4 in number. Zone 1 reach setting is often set at 80% of the
protected line and that of zone 2 is at 100% of the protected line, plus 50% of
the shortest next line. Zone 3 setting is set to cover 100% of the protected line,
and 100% of the second longest line, plus 25% of the shortest remote line. Zone
1 operation very often is instantaneous and operating times for Zones 2 and 3
are typical in the ranges of 0.25-0.4 s and 0.6-1.0 s, respectively. Distance
protection relays have characteristics to define its protected region. Mho and
quadrilateral are two of the most commonly employed distance characteristics.
In some instances, reverse looking zones are also defined.
Figure 8: Distance characteristics on R-X (resistance-reactance) plane for three zones; (a)
Mho, (b) Quadrilateral (line impedance is plotted in dashed line)
2.2.2 Challenges in Protecting Series Compensated Transmission Lines
Capacitor banks are often connected in series with transmission lines in
order to increase their power transfer capabilities. Protection of these series
compensated transmission lines is one of the long prevailing challenges in
X X
Zone 3
Zone 2
Zone 1 Zone 2
Zone 1
Zone 3
R R
(a) (b)
31
power system protection [19]. The occurrence of voltage and current inversions
in a series compensated transmission line is explained below with the help of
Figure 9.
Figure 9: Voltage and current inversion in a series compensated transmission line
Voltage and current inversions across the series capacitor can provide
complications to conventional protection algorithms and make traditional
distance relays to mal-operate. When the series capacitor is in service, the bus
side voltage, VB, is equal to I. (X𝐿,𝑓 − X𝐶). j and the line side voltage, VL is equal
to I. X𝐿,𝑓. j. For certain faults, where the inductive reactance of the faulty line
section is smaller than the capacitive reactance of the series capacitor (X𝐿,𝑓 <
X𝐶), VB gets inverted by 180° [19]. A distance relay, fed with VB, sees such a
fault (which should be a zone 1 fault in most cases) as a reverse fault and will
restrain itself from operating. This can be avoided by feeding the relay with
the line side voltage, VL. However, then voltage inversion happens for reverse
faults. The relay will consequently see certain reverse faults as forward and
could mal-operate.
Current inversion happens when the impedance between the power
system source and fault point is capacitive (XS + XL,f < XC). In this case, the
XL, f
I
VL
XC
VB
XS
32
current at bus 1 would be capacitive, while current at bus 2 is still inductive.
This causes a complication in protection algorithms as the phase relationship
of the two currents in this scenario is similar to that of an external fault [19].
Furthermore, sub-harmonic frequency transients in series compensated lines
can induce oscillations in impedance estimations of the relays, causing them
to overreach.
2.2.3 Current Transients Based Line Protection
Analysis of high frequency electromagnetic transients caused by various
power system events is a well-known strategy in power system protection.
High frequency transients, particularly those superimposed in line currents,
are commonly analysed in modern protection algorithms [20]-[25]. These
protection schemes usually employ time-frequency-transformations to extract
the high frequency transients generated by a power system abnormality such
as an occurrence of a fault or a switching operation. The extracted waveforms
are further examined to gain information about the event including its location
and magnitude.
2.2.3.1 Filtering methods used for transient extraction
There exist several means of digital signal processing for the extraction
of high frequency transients superimposed in line currents and voltages. The
main methods are conventional filtering, short-time Fourier transform (STFT)
33
and wavelet transform (WT). The use of the wavelet transform in power system
studies is relatively newer than the other two techniques.
Regular band-pass filters can be employed for the study of high
frequency transients. However, WT has several advantages over conventional
filtering, including better performance in signal de-noising [21]. In addition,
the selection of the optimum threshold frequencies for the filters can be a
complication, when using conventional filters.
Conventional Fourier Transform cannot be used to analyze momentary
high frequency components in line currents, as it does not provide any time
localization. However, both STFT and WT deliver frequency decomposition of
a signal and provide means of two dimensional localization of time-frequency.
The main difference between the two techniques is the window size of STFT is
fixed, in contrast to that of WT, which is scalable (as a result, the frequency
resolution of STFT is uniform and, on the other hand, WT has multi-resolution
properties) [20].
The transients based algorithm proposed in this work depends on the
capability of the signal processing unit to deliver the polarity of the wave-front
of a high frequency transient. With STFT, all information about the high
frequency transient (contained within the considered time window) is available
in the frequency domain. Hence, obtaining information related to the wave-
front polarity is not straightforward. In contrast, WT extracts the required
frequency band in time domain, and as a result, the polarity of the wave-front
34
is readily available. The WT is, therefore, better suited for the application
considered in thesis and a brief explanation about WT and its implementation
methods are provided below.
2.2.3.2 Wavelet Transform
The wavelet transform of a continuous signal x(t) is given by [20],[26],
WT(a, b) =1
√a∫ x(t). g (
t − b
a) dt
∞
−∞
; Where, g(t) is referred to as the “mother wavelet” and “a” and “b” are scaling
and translation constants, respectively. For a function to be considered as a
mother wavelet, it must have zero average and be quickly decaying at each
end. By definition, an infinite number of mother wavelets can exist, but there
exist several commonly used families of mother wavelets. The selection of the
mother wavelet for a certain study is governed by the requirements and the
nature of the study. Two example mother wavelets are given in Figure 10.
Figure 10: Wavelet function of (a) ‘db4’ mother Wavelet, (b) ‘Symlets 2’ mother wavelet
0 1 2 3 4 5 6 7 8
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1
1.2db4 : phi
0 1 2 3 4 5 6 7 8-1
-0.5
0
0.5
1
1.5
(a)0 0.5 1 1.5 2 2.5 3 3.5
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1
1.2
1.4sym2 : phi
0 0.5 1 1.5 2 2.5 3 3.5-2
-1.5
-1
-0.5
0
0.5
1
1.5
(b)
35
Moreover, the discrete form of the wavelet transform is referred to as
the discrete wavelet transform (DWT) and can be given as [20],
DWT(m, n) =1
√a0m
∑ x[n]. g [k − na0
m
a0m
]
k
It has been mathematically proven that the DWT can be implemented
using a series of low and high pass filters as shown in Figure 11 [20], [26].
Figure 11: Multi-level filter bank implementation of DWT; (2↓) denotes down-sampling by 2
Here, the signal x[n] is decomposed using a low pass filter, g[n] and its
high pass dual, h[n]. They are known as quadrature mirror filters and their
filter coefficients are the reverse of each other, with every other coefficient
negated [20]. Further, the outputs of the high pass filter are called “detail
coefficients”, whereas those of the low pass filter are called “approximation
coefficients”.
These filters are designed in such a way that their output has half the
frequency band of that of the input. Since the top half of the frequencies are
removed by h[n], one half of the samples can be dispensed with before the next
Level 3 detail
Coefficients
Level 2 detail
Coefficients
Level 1 detail
Coefficients
x[n]
h[n] 2 ↓
2 ↓ g[n]
h[n] 2 ↓
2 ↓ g[n]
h[n] 2 ↓
2 ↓ g[n]
36
level, as per Nyquist–Shannon sampling theorem. The effect of down-sampling
the filter output by 2 is equivalent to the scaling of the wavelet function by a
scaling factor of 2 in the subsequent level [20]. This scheme can be
implemented using any wavelet after determining the corresponding filter
coefficients. Filter coefficients for commonly used wavelets are found in [27].
Chapter Summary
In this chapter, a comprehensive introduction to the standard IEC
61850 was provided with its advantages, definitions, architectures and
protocols. The emphasis was given on sampled values (SV) protocol. Further,
an overview of transmission line protection using impedance based methods
was provided. Basics of transients based protection were also briefly explained.
Chapter 3
Implementation of a Process Bus
Compatible IED Using Generic
Hardware
Experimental Setup
At the beginning of this section, the basic arrangement of the
experimental setup, including the relay developed in this research, is
presented in order to provide an overall picture about the function of the
developed system. This is followed by detailed explanations on the
implementation of the line protection IED application.
The power system, on which the developed IED is tested and validated,
is simulated in RTDSTM. It is a real-time digital simulator based on
electromagnetic transient (EMT) simulation (this simulator, form this point
onwards, shall be referred to as the RTDS). It is also the primary SV source
considered in this work. Depicted in Figure 12, is the schematic of the
38
experimental setup (related to the distance protection scheme) developed for
this research. The dashed-box in Figure 12 represents the implemented IED
that is existing outside the RTDS on a desktop computer. Real IEC 61850
communication is used for transferring information between the RTDS and the
IED. All other components are modelled inside the RTDS simulation case,
using its standard library modules. The power system section under
consideration is an overhead transmission line connecting two AC networks.
Figure 12: Schematic of the apparatus for the distance protection scheme
Implementation of the IEC 61850 substation architecture and
communication system is also represented in Figure 12. The network interface
GOOSE (trip) Protection IED
(on computer)
MU (GTNET1)
GTNET2
Circuit
Breaker IED
VT CT
Fault
Process
Bus
Transmission line
L
o
a
d
Sampled Values
RTDSTM
Bus 1
Switch
Switch
Bus 2
39
card of the RTDS, referred to as the GTNET card [28], supports the two
protocols GOOSE and SV. For this work, two such GTNET cards are used, one
for each protocol.
Sampled values (SV module) module in the RTDS is compliant with IEC
61850-9-2 LE, thus, publishes four currents and voltages each to form a virtual
MU inside the RTDS. Detailed modelling of this MU will be explained later.
This SV module is connected with GTNET1, which is dedicated for SV
communication. In this study, the published SV streams are the instantaneous
currents and voltages of the three phases and the neutral, measured at Bus 1,
as shown in Figure 12.
The SV steam published by GTNET1 goes through a high speed (1 Gbps)
Ethernet network (the process bus) into the line protection IED (implemented
on the desktop computer), where it is subscribed and decoded to be used in
protection algorithms. Once the relay detects a disturbance by internal
processing of the received information, it issues a trip signal via a GOOSE
message. This GOOSE message is routed through a separate 100 Mbps
network and is subscribed by the second GTNET card (GTNET2) of the RTDS.
The GOOSE module in RTDS simulation then decodes the information inside
the message and acts accordingly by tripping the circuit breaker in the
simulated network.
Moreover, form here onwards, the near end refers to Bus 1 and the far
end refers to Bus 2 of the transmission line, as shown in Figure 12.
40
Furthermore, all distances (for instance, distance to the location of a fault) are
measured from Bus 1.
Line Protection IED Application
The basic line protection relay (IED) application developed in “C”
programming language runs on desktop computer with a Linux operating
system. Basic implementation structure of the developed IED is illustrated in
Figure 13 below. The IED application (the line protection relay), once
configured and initiated, is looking for a source that is publishing SV
(according to IEC 61850-9-2-LE). The process of matching IEC 61850
configurations of the source with those of the destination is called mapping,
which will be discussed under a different topic later.
Once a connection is established with a SV source (in this scenario, with
the RTDS), the application starts receiving SV. These SV packets are then
stored in a circular data buffer. Concurrently, another tread of the application
reads SV packets from the same buffer and then, decodes them to acquire the
instantaneous currents and voltage values contained within. These values are
then subjected to further processing as shown in Figure 13.
As the next step, any DC offsets contained within the currents and
voltages are filtered-out using a digital mimic filter (a low pass filter).
Afterwards, discrete Fourier transform is performed using this filtered data,
in order to extract the current and voltage phasors. Impedance estimations are
41
carried out using the said phasors, which are then fed into the distance relay
characteristics. If the estimated impedance is within a zone, a trip signal is
issued via a GOOSE message. Important sections of the above
implementations are explained in detail in the next sub-sections.
Figure 13: Implementation structure of the line protection IED; only the distance protection
function is illustrated
Filtered
Instantaneous
values for
phasor
estimations
GOOSE
Va
SV
packets as
per IEC
61850-9-2
LE
Read SV
buffer,
decode
Voltage and
Current
samples
Mimic Filter:
Removal of
DC offset
Subscribe SV
packets, fill
the buffer
SV
buffer
Vb
Vc
Vn
Ia
Ib
Ic
In
Va
Vb
Vc
Vn
Ia
Ib
Ic
In
Va
Phasor
Extraction:
Perform
Discrete
Fourier
Transform
Vb
Vc
Vn
Ia
Ib
Ic
In
Distance
estimation
units, by
Phase and
ground
units
Distance relay
characteristics,
Mho or
quadrilateral
If a fault is
detected,
initiate a trip
signal and
publish a
GOOSE
42
3.2.1 Implementation of IEC 61850 Services
This application uses a set of software libraries developed by Kalkitech
Pvt. Ltd, India, to accomplish the basic IEC 61850 services such as publishing,
subscribing, storing and decoding of SV and GOOSE messages. The Kalkitech
IEC 61850 library consists of six modules:
1. Initialization Module.
2. Server Module.
3. Goose Publisher Module.
4. Goose Subscriber Module.
5. Sampled value Subscriber Module.
6. Sampled value Publisher Module.
Each of these modules contains different sets of APIs (Application
Programming Interface) for IEC 61850 services, which are listed in Appendix
A-I. This high level API library is built on SISCO SCL MMS Ease Lite [29],
which is a source code package for the Manufacturing Message Specification
(MMS) protocols, IEC 61850 that has been optimized for use in IEDs such as
RTUs, reclosers, PLCs, meters and other resource constrained embedded
applications. However, several of the functions in Kalkitech IEC 61850 library
needed to be modified to achieve the desired functionality in this IED
application.
43
3.2.2 Digital Mimic Filter
Depending on factors such as fault inception angle and X/R ratio of the
network, current and voltage waveforms following a fault may contain
decaying DC offsets. These DC offsets cause the phasor estimations (obtained
using DFT) to be erroneous by inducing small oscillations on them. To
eradicate this problem, a high pass filter is often used in modern digital relays
and it is referred to as the “mimic filter”.
It can be proven from basic circuit theory that if an AC signal with an
exponentially decaying DC component is passing through a filter with the
transfer function,
H(s) = K(1 + sτ)
Where, τ is equal to the time constant of the decaying DC component, the
output waveform exhibits no DC component. The discrete version of the above
transfer function can be written as [30],
H(z) = K(1 + τ) − Kτz−1
Applying the mimic filter to a discrete sinusoidal signal x(k) given by,
x(k) = A1e−k.Δt
τ + A1sin [(2π. Δt. k
W) + ϕ]
Where, W is the number of samples per cycle and Δt is the sampling interval.
Applying the mimic filter to x(k) would yield,
44
y(k) = K[(1 + τd). x(k) − τd. x(k − 1)]
In (6), τ is expressed as a number of samples (τd). It can be proven that for the
fundamental component, the gain K is given by [30],
K =1
√((1 + τd) − τd. cos (2πW
))
2
+ (τd. cos (2πW
))
2
Further, this mimic filter introduces a phase shift to the AC signal which can
be expressed as,
ψ = tan−1 [τd. cos (
2πW
)
(1 + τd) − τd. cos (2πW
)]
Implementation of the mimic filter for the IED is carried out using the
above design. There, the number of samples per cycle, W, is 80 and τd is a
setting dependent on the power system X/R ratio. For this work, this user
settable parameter is set at 640 samples (8 cycles). On the other hand, it is also
important to know that this filter introduces a delay of 1 sample (equivalent to
a time delay of 0.20833 ms at 80 samples per cycle) to the phasor estimation
process. An example of a current waveform with a dc offset before and after the
mimic filter is given in Figure 14. These sample results obtained with τd = 640
samples (8 cycles) shows the effectiveness of the mimic filter.
45
Figure 14: (a) Unfiltered current waveform, (b) Filtered current waveform
7.35 7.4 7.45 7.5 7.55 7.6 7.65-5
0
5
10
Phase A Current waveform (unfiltered)
Time (s)
Cu
rre
nt (k
A)
7.35 7.4 7.45 7.5 7.55 7.6 7.65
-6
-4
-2
0
2
4
6
Phase A Current waveform (filtered)
Time (s)
Cu
rre
nt (k
A)
(a)
(b)
46
3.2.3 Phasor Extraction using Discrete Fourier Transform
A number of algorithms can be used for phasor estimation in power
systems. Commonly used algorithm classes are Discreet Fourier Transform
(DFT) based methods, optimization based methods such as Least Square
Estimation, wavelet based methods and dynamic phasor based methods [31].
DFT based algorithms are simple and widely employed in modern digital
relaying for phasor estimations, and hence, used in this work to extract the
phasors from sampled data.
From the relationship between the Discreet Fourier Transform and the
Fourier series, the fundamental frequency component extracted from N
number of samples of the given waveform, x(t), can be written as,
X1 =√2
N∑ x(n. Δt). cos (
2πn
N)
N−1
n=0
− j.√2
N∑ x(n. Δt). sin (
2πn
N)
N−1
n=0
Where, Δt is the sampling interval. In power system protection, current and
voltage waveforms having a nominal frequency of f0, are assumed to have the
general format,
x(t) = Xmcos (2πf0t + ϕ)
When the signal is sampled at W samples per cycle (a sampling frequency of
W.f0), and ϕ is the phase angle between the first sampling point and the peak
of the signal, it can be written as,
47
x(n) = Xmcos (nθ + ϕ)
X1Re =√2
N∑ x(n). cos(nθ)
N−1
n=0
and X1Im =√2
N∑ x(n). sin(nθ)
N−1
n=0
Where, θ=2π/W is the phase angle corresponding to the sampling interval.
Then the fundamental phasor (denoted by subscript 1) is given by,
X1N−1 = X1Re − j. X1Im
When using the recursive algorithm to extract the phasor, the general form of
the real and imaginary parts of the fundamental phasor can written as, (where
(N+r) is the last sample of the window)
X1ReN+r = X1Re
N+r−1 +√2
N(𝑥𝑁+𝑟 − 𝑥𝑟). cos(rθ)
X1ImN+r = X1Im
N+r−1 +√2
N(𝑥𝑁+𝑟 − 𝑥𝑟). sin(rθ)
This can be expressed in the following format as well.
X1Re(K) =√2
W∑ x(n). cos(nθ)
K−1
n=K−W
X1Im(K) =√2
W∑ x(n). cos(nθ)
K−1
n=K−W
X1(K) = X1Re(K) − j. X1Im(K)
Where, W is the sampling window size and (K-1) is the last sample of the
sampling window (with the first and the last sample numbers being 0 and (N-
1) for N samples). This means K increases, starting from the sampling window
size up to the number of samples, N.
48
The relevant parameters for phasor extraction using the DFT in the IED
are given below.
f0 = 60 Hz, W = 80 samples/cycle, W. f0 = 4800 Hz
Δt, sampling invertal =1
4800= 0.00020833 s
Here, the DFT is performed using a sliding data window of 80 data
points, which is implemented by a dynamic data structure called “linked list”.
An example phasor estimation performed in the IED is given in Figure 15
below. Note that the algorithm takes one sample window, in this case 80
samples, to settle down at the new value after a change. Plotted in the top most
graph is the instantaneous values of a current, obtained from the SVs received
by the IED.
49
Figure 15: Magnitude and phase of the extracted current phasor plotted against the sample
number
300 400 500 600 700 800 900 1000 1100 1200 1300-1500
-1000
-500
0
500
1000
1500
X: 615
Y: 264In
sta
nta
ne
ou
s C
urr
en
t (A
)
300 400 500 600 700 800 900 1000 1100 1200 1300100
200
300
400
500
600
700
X: 615
Y: 161.1
Cu
rre
nt M
ag
nitu
de
(A
)
X: 696
Y: 659.5
300 400 500 600 700 800 900 1000 1100 1200 13000
0.1
0.2
0.3
0.4
0.5
X: 615
Y: 0.1241
Sample Number
Cu
rre
nt p
ha
se
(ra
d)
X: 696
Y: 0.4137
50
IEC 61850 Communication Configurations
Specific configurational procedures should be carried out before
commencing communication according to IEC 61850 standard. A process called
mapping is of particular importance, where entities such as published
variables and data objects of the publisher are mapped to appropriate inputs
of the subscriber. Moreover, network parameters such as MAC addresses and
VLAN IDs should also be properly configured.
Since the communication in this work is bidirectional, data mapping
should be carried out accordingly. That is, the data mapping should account
for both the SV communication from the RTDS to the developed IED and the
GOOSE communication from the developed IED to the RTDS. General
procedures for publication and subscription of information in IEC 61850 are
explained in the upcoming sub-sections in order to provide a comprehensive
picture of the information exchange occurring in this work.
3.3.1 Publication and Subscription of Information in IEC 61850
The publication and subscription of information in IEC 61850 are
accomplished by following rigorous procedures. According to IEC 61850, all
engineering design and data flow configuration in a project should be carried
out in the system configuration tool, employing a unified approach [32].
However, the use of a system configuration tool for overall system
51
configurations is not well established in existing commercial IEC 61850
implementations. The prevailing norm is to use the proprietary (vendor-
specific) IED configuration tools to setup the IEDs. Typically followed steps for
setting up an IED for IEC 61850 based communication are explained below
(which is also the method employed in this research as well).
3.3.1.1 Publication Procedure
Firstly, the entities required to be published are put together into a
group, which is called as a “dataset”. These entities are often Functionally
Constrained Data Attributes (FCDA), although other entities in IEC 61850
product hierarchy such as logical nodes and data objects can also be published.
Any number of datasets can be created as the requirements of the project
demand. Then a control block is created including a particular dataset, and
this control block contains additional information such as MAC address,
application ID, VLAN (virtual local area network) ID and VLAN priority. A
particular message published according to IEC 61850 is governed by its control
block. This procedure of publication is common for both GOOSE and SV, even
though their message initiation is different from each other. i.e., a new GOOSE
message is initiated by a GOOSE control block once it detects a change in any
of its dataset variables, whereas a SV control block publishes a new SV packet
with all of its dataset variables, periodically.
52
Above mentioned configuration steps are carried out in the ICD file of
the publisher’s IED configuration tool. That file is then saved as a CID file and
loaded to the respective publisher IED. This same file is also exported into the
subscriber’s IED tool for subscription configurations.
3.3.1.2 Subscription Procedure
In contrast to the publication process, the method of subscription of
messages in IEC 61850 is not entirely consistent. The method of subscription
defined in IEC 61850 is using the Input/ExtRef element, which enables the
binding of the external signals to an internal address of the IED. The nature
of existing IEC 61850 implementations by almost all IED vendors is such that
data mapping can only be performed with the subscriber’s own IED
configuration tool.
In the usual procedure, the CID files of all the publisher IEDs are
imported into the subscriber’s IED configuration tool, and required publisher
variables are mapped into the internal variables of the subscriber IED. All
these configuration information is contained in the subscriber’s CID file, which
is saved and, subsequently loaded into the corresponding IED. Moreover, the
syntax for signal mapping in subscription is not defined in IEC 61850. As a
result, syntaxes used vary from vendor to vendor and these are found enclosed
inside private (<private>) sections in the subscriber’s CID files.
53
3.3.2 Network Interface Card of the RTDS
Network Interface Card (NIC) of the RTDS, which is named “GTNET”
card, supports five communication protocols in total, four of which can be kept
installed through firmware. Moreover, only one chosen protocol of those four is
functional at a given point of time. The functionality of the GTNET card has
two aspects to it. Firstly, it can accept packets from a LAN, then, extract the
information inside those packets and send it to the processor card (called PB5
card) of the RTDS to be used in the simulation. Secondly, it can get the values
from the simulation, encode them into packets and send them out to be
subscribed by other devices [28]. The list of supported protocols by the GTNET
card are given below,
GSE: IEC 61850 GOOSE/GSSE
SV: IEC 61850−9−2 (Sampled Values)
DNP3: Distributed Network Protocol
PLAYBACK: Playback of large data sets
PMU: IEEE C37.118 datastream output
The RTDS simulator used for this work has two GTNET cards, with
GOOSE and SV as active protocols. A GTNET card has its own MAC address
and IP address. In addition to the GTNET and the PB5 cards, few of the other
card types in the simulator RTDS are also of significance to this study. One
54
such card is GTSYNC or synchronization card. Its purpose is to make sure that
the RTDS simulation is in synchronism with an external, high precision time
reference, which could be one of IEEE 1588 PTP (Precision Time Protocol), 1
PPS, or IRIG−B signals. For this research, the GTSYNC is provided with an
IRIG−B signal, taken from a clock that is synchronized to a GPS signal. Clock
used here is a SEL-2407, Satellite Synchronized Clock.
All Physical connections in the setup are shown in Figure 16. Connections
among different cards inside the RTDS are shown in dashed lines.
Figure 16: Physical connections in the LAN
Ethernet switch 2 Ethernet switch 1
IRIG−B Signal from
the clock
RTDS
Ethernet LAN
Line protection IED
residing in a desktop
computer
GTNET 1
(SV)
GTNET 2
(GOOSE)
PB 5
(Processor)
SV
GOOSE
GTSYNC
GTWIF
To computer workstation
running the RSCAD software
55
GTWIF or the Workstation Interface Card of the RTDS manages all
other cards in the simulator. It communicates with RSCAD, which is the
software interface of the RTDS, in order to start/stop simulations and to
exchange information regarding plotting, user initiated events etc. [28]. The
connections of the Ethernet LAN, carrying GOOSE and SV between the RTDS
and the developed line protection IED are also shown.
3.3.3 SV Publication from the RTDS
SVs in the RTDS are handled by the sampled value component in the
RTDS simulation, as shown in Figure 17. It can be configured to either publish
or receive SV and it supports both 9-2 LE and non-LE protocols. Here, non-LE
refers to a protocol developed based on IEC 61869−9 and the Chinese national
standard for SV merging units. In this work, the Sampled Value component is
used only for SV publication, under 9-2 LE configuration, with a sampling rate
of 80 samples per cycle (while 256 samples per cycle rate can also be used).
As depicted in Figure 17, the secondary three phase current and voltage
signals of the two instrument transformers (IA, IB, IC and VA, VB, VC) are fed
to a series of anti-aliasing low pass filters. A series of 6th order low-pass
Butterworth filters is used for anti-aliasing. The cut-off frequency of these
filters is set at 2.4 kHz, which is 50% of the sampling frequency of 9-2 LE (with
the system frequency at 60 Hz, 80 samples per cycle is equivalent to 4.8 kHz).
The filtered signals are then taken by the sampled value component to be
56
published as SVs through GTNET hardware. Furthermore, the combination of
the anti-aliasing filters and the sampled values component can be considered
as a virtual merging unit as per IEC 61850-9-2 LE. This margining unit is
indicated in Figure 12 in above Section 3.1. Also notice that neutral current
and voltage are calculated using the phase values.
Figure 17: Anti-aliasing filters (left) and the sampled values component (right), residing in
the RTDS simulation case
RTDS creates its own CID (the RTDS uses the extension .icd for CID file
type as well, however to avoid confusion, it is referred to as the RTDS CID file)
file for IEC 61850 communication configurations. This CID file contains a
dataset, including eight FCDAs. I.e. four FCDAs from current transformer
(CT) logical nodes (ex: TCTR1.MX.Amp.instMag.i) and four FCDAs from
voltage transformer (VT) logical nodes (ex: TVTR1.MX.Amp.instMag.i). As
57
explained in 3.3.1.1, this CID file has a SV control block to manage the
publication of the said dataset. The CID file for SV publication of the RTDS is
provided in Appendix A-II.
3.3.4 SV Subscription by the Line Protection IED
The ICD file for a given IED is provided by the manufacturer. For the
line protection IED developed in this research, an ICD file is created manually,
starting from scratch, using a commercial IEC 61850 SCL file editing software
tool (KalkitechTM SCLManager) with all the required logical node instances
and data type definitions. Since there is no IED configuration tool for the
laboratory developed IED, this ICD file is manually configured to create the
CID file of the line protection IED. SV subscription in the IED is achieved by
means of virtual current and voltage transformer LN instances. Here, the word
“virtual” is used to describe the LN instances because they do not represent
any physical (existing) elements in the substation (unlike the ones in the RTDS
publisher CID file, which represent the actual current and voltage transformer
of the substation). Notwithstanding, the advantage of this method is that
subscription of SVs is only done at a single place of the IED and the local
implementations of LN functions can access the subscribed data internally,
without having to subscribe SVs individually to them. The current and the
voltage transformer LN instances in the CID file of the IED are exact replicas
58
of their counterparts in RTDS CID file, thus, the signal mapping is one to one.
This is illustrated in Figure 18 below.
Figure 18: Signal mapping for SV subscription
TCTR1.MX.Amp.instMag.i
TCTR2.MX.Amp.instMag.i
TCTR3.MX.Amp.instMag.i
TCTR4.MX.Amp.instMag.i
TVTR1.MX.Amp.instMag.i
TVTR2.MX.Amp.instMag.i
TVTR3.MX.Amp.instMag.i
TVTR4.MX.Amp.instMag.i
RTDS CID file
TCTR1.MX.Amp.instMag.i
TCTR2.MX.Amp.instMag.i
TCTR3.MX.Amp.instMag.i
TCTR4.MX.Amp.instMag.i
TVTR1.MX.Amp.instMag.i
TVTR2.MX.Amp.instMag.i
TVTR3.MX.Amp.instMag.i
TVTR4.MX.Amp.instMag.i
Line protection IED CID file
59
3.3.5 GOOSE Publication from the Line Protection IED
Upon detection of a fault, the developed protection IED sends a GOOSE
message to the circuit breaker simulated in the RTDS. For issuing of trip
signals in the distance protection scheme, instances of the LN class “PDIS”
(Distance Protection logical node) are used. However, there is no standard LN
class available for transient based protection scheme, since it is a user defined,
proprietary protection scheme. Therefore, instances of the LN class GGIO
(Generic Process Input/Output) are used for this purpose.
3.3.6 GOOSE Subscription by the RTDS
In a RTDS simulation case, there can be many devices (circuit breaker
controllers, protection relays, etc.) those use the information in GOOSE
messages sent by external IEDs. However, since communication is handled by
the GTNET cards, configuration of all IEC 61850 communication services is
achieved through a common CID file that treats RTDS as a single IED.
Information inside any GOOSE message received by a GTNET card is decoded
by the relevant GTNET-GSE block in the RTDS simulation case and then
assigned to an internal variable, as specified in the parameter section of the
GTNET-GSE block. This internal variable usually is an input or an internal
variable in a device simulated in RTDS. For GOOSE messaging, the RTDS only
supports the LN class GGIO. One GTNET-GSE block in the RTDS can
accommodate up to 64 GOOSE input/outputs, although the number of inputs
60
used here are 1 or 2, depending on the protection scheme in use. GOOSE
messages published by the line protection IED are subscribed by the GTNET
hardware of the RTDS as explained, and this is illustrated in Figure 19.
Figure 19: An example signal mappings for GOOSE subscription in; (a) Distance Protection
scheme (b) Transients based protection scheme
Given in Figure 19 are two examples of signal mappings for GOOSE
subscription for the two protection schemes. Note that the all published
variables are binary values. Furthermore, for the distance protection scheme,
the GTNET hardware and the GTNET-GSE component in the RTDS, forms
what is referred to as a “breaker IED” (BIED) in the virtual environment of the
simulation case. A BIED is the smart section of a circuit breaker (recognized
as an IED in the perspective of IEC 61850) that controls its operation in a
substation according to the control signals provided by the protection schemes
or an operator. These devices reside in the process level and, are connected to
the circuit breakers using hardwires internally.
PDIS1.ST.Op.general
RTDS CID file
GGIO1.ST.Ind1.stVal
Line protection IED CID file
GGIO1.ST.Ind1.stVal
RTDS CID file
GGIO1.ST.Ind1.stVal
Line protection IED CID file
(a)
(b)
61
Figure 20 illustrates the implementation of a breaker IED in the RTDS
simulation. The decoded binary value of the subscribed GOOSE message (the
trip signal) is assigned to the variable BRK, the value of which determines
position of the circuit breaker (opened or closed).
Figure 20: Illustration of the implementation of a breaker IED in the RTDS simulation
Time Delays in the Setup
The time taken from the implemented line protection IED to respond to
a fault simulated in the RTDS is very large compared to the response time of
actual relays. This time delay is measured as a round trip time (time take from
the moment a fault is applied in the RTDS simulation to the arrival of the
GOOSE trip from the relay), which is observed to be varying in rather a wide
range from 10 ms to 2 s. This round trip time includes the processing times in
the RTDS, communication network delays and the processing time in the
implemented line protection IED.
Incoming GOOSE message
GTNET card
BRK
GTNET-GSE
component
Circuit breaker component
BRK: variable controlling the circuit breaker operation
Breaker IED
62
It is found out that a major portion of this round trip time
(approximately 95% of it) is caused between the arrival of SV packets at the
IED (network card of the desktop computer) and passing of the decoded
sampled values to signal processing modules. Closer examination showed that
there is no SV packet loss, but just a random delay that is not accumulating.
Although the exact reason behind this problem is not identified, it is assumed
that the delays in the network interface card of the desktop computer and the
non-real-time nature of its operating system are major contributing factors to
the delays. However, irrespective of the delay, IED responds correctly to every
fault. Due to this time delay, the time performance of the implemented line
protection IED is not evaluated.
Chapter Summary
Chapter 3 rigorously explained the basic implementation of the line
protection IED application. Its internal algorithms for filtering and phasor
calculations were provided with necessary equations and algorithms. Some
results were also provided for validation of the filtering and the phasor
calculation. Moreover, all IEC 61850 communication related information,
including publishing and subscribing procedures at both ends, were described
in detail.
Chapter 4
Distance Protection Scheme
Distance protection scheme developed in the line protection IED has
both Mho and quadrilateral characteristics at its disposal. The protection
scheme is tested and verified on the power system simulated in the RTDS.
Implementation details are described in the next few sections.
Distance Estimation Units
A distance protection relay has two types of distance estimation units;
namely, phase units and ground units. Usually a distance protection scheme
requires six units (three each from the two types), for it to identify all types of
faults involving all three phases. The phase units identify three-phase (3-Φ)
faults and line to line (L-L) faults, whereas the ground units identify line to
ground (L-G), line to line to ground (L-L-G) and three-phase to ground (3-Φ-G)
faults. Moreover, independent voltage and current inputs are provided for each
of these six units. Well-known equations for impedance estimations from
64
calculated voltage and current phasors, in a phase unit (AB) and a ground unit
(AG) are given below.
ZAB = VAB
IA − IB=
VA − VB
IA − IB
ZAG = VA
IA + 3K0I0 ; where K0 =
Z0 − Z1
3Z1
Here, ZAB and ZA are impedances seen by the phase unit and the ground
unit, respectively. VAB is the line to line voltage between phases A and B, while
VA and VB are corresponding line to neutral voltages. Similarly, IA and IB are
line currents of the two phases, whereas I0 is the zero sequence current. Z1 and
Z0 are, respectively, the positive and the zero sequence impedances of the
protected line. K0 is referred to as the zero sequence compensation factor. This
K0 is introduced in such a way that the impedances seen by the ground units
also are the positive sequence impedance of the line (as is the case with the
phase units).
In the implementation of the laboratory line protection IED (in
computer application code), two functions are developed to calculate above
phase unit and ground unit impedances, by taking the extracted phasors as
arguments. For instance, calling of the phase unit function with arguments VA,
VB, IA and IB would yield, ZAB; the impedance seen by the phase unit AB.
Likewise, the two functions are called three times each, in order to determine
all six impedance estimations. These impedance estimations are then passed
as arguments to another function implementing the impedance characteristics.
65
Impedance Characteristics
4.2.1 Mho Characteristic
The impedance setting of the zone-1 shown in Figure 21 is at 80% of the
positive sequence impedance of the protected transmission line, denoted as Z1
(Z1∠Φ). The angle Φ is the impedance angle of the line.
Figure 21: Implementation of zone 1 Mho characteristic
Notice that impedance line, AB, passes through the centre of the Mho
circle on the R-X plane. Furthermore, the origin and the radius of the Mho
circle are denoted as O and r, respectively. The x and y coordinates of the
origin, O and the magnitude of the radius, r are expressed in terms of Z1 and
Φ as follows,
C
B
A
O
From the impedance
estimations
Zest = Zre, Zim
Zone 1 Mho characteristic
with a forward reach
setting of 0.8Z1 r'
jX
Φ
Z1 Zest = Zre, Zim
r
R
If r’ ≤ r ⇒ Zone 1 fault
66
O = (x:0.8Z1
2cos(Φ) , y:
0.8Z1
2sin(Φ))
r =0.8Z1
2
The distance from O to the point representing the estimated impedance Zest is
given as r’, and it can be calculated in terms of x, y, Zre and Zim as,
r′ = √(x − Zre)2 + (y − Zim)2
Where Zre and Zim are, respectively, the real and imaginary parts of the
estimated impedance point, Zest. It is clear that the estimated impedance point
resides within the Mho characteristic circle, if (and only if) the radius of the
circle is greater than the distance between its origin and the impedance point.
The fulfilment of this condition, therefore, indicates that a zone 1 fault has
been occurred. Required number of zones can be obtained by varying the
radius of the Mho circle and extending a similar approach.
The forward reach (length AC in Figure 21) in secondary Ohms is the
typical setting for zones 1 and 2 in a distance protection scheme using Mho
characteristics. Zone 3 and beyond could incorporate a reverse reach as well.
Further, the characteristic angle, Φ, can also be a setting in some schemes.
67
4.2.2 Quadrilateral Characteristics
Zone 1 quadrilateral characteristic on R-X plane is depicted by the
quadrilateral “abcd” in Figure 22. Zest, Z1 and Φ have same meaning as in the
previous sub-section.
Figure 22: Implementation of zone 1 quadrilateral characteristic
Typically, a quadrilateral characteristic has three settings. They are
forward reach (length AC), right reach (Rr) and left reach (Rl). Gradients and
the intersections of the four straight lines that form the quadrilateral “abcd”,
can be determined using the aforementioned setting as follows,
k = 0.8𝑍1𝑠𝑖𝑛(Φ)
m1 = 𝑡𝑎𝑛(Φ)
m2 = 𝑡𝑎𝑛(Φ + 𝜋 2⁄ )
C d c
b
a
𝐲 = 𝐦𝟐𝐱
𝐲 = 𝐤
𝐲 = 𝐦𝟏𝐱 + 𝐜𝟐
𝐲 = 𝐦𝟏𝐱 + 𝐜𝟏
Rl Rr
B
A
From the impedance
estimations
Zone 1 quadrilateral
characteristics with a forward
reach setting of 0.8Z1
jX
Φ
Z1
Zest = Zre, Zim
R
Zest = Zre, Zim
68
c1 = − [𝑅𝑟𝑡𝑎𝑛(Φ) +𝑅𝑟
𝑡𝑎𝑛(Φ)]
c2 = [𝑅𝑙𝑡𝑎𝑛(Φ) +𝑅𝑙
𝑡𝑎𝑛(Φ)]
As before, the criterion for fault identification is to check whether the
estimated impedance point lies inside the defined zone. This is occurring when
the following four conditions are satisfies simultaneously.
Zim > m1Zre + C1
Zim < m1Zre + C2
Zim < k
Zim > m2Zre
And similar to Mho characteristics, the other zones can be implemented
by simply increasing above mentioned the reach settings.
69
Fault Identification
Both types distance characteristics are implemented in a single function
that takes estimated impedance as the input. Only one type of characteristic
can be in use at a given time. This functions returns an indicator (with value
1) when it sees the estimated impedance point inside the predefined zone.
Fault identification algorithm for the distance protection scheme is given in
Figure 23.
Figure 23: Fault Identification algorithm of the distance protection scheme
The IED is going through this process for each SV packet received. The
algorithm demands that 20 consecutive impedance estimation points must be
Estimated
impedance point
NO YES Inside
Zone
Extracted phasors
Next packet
Impedance estimation
Distance characteristic
Counter ++ Reset counter
NO YES Counter > 20
Next packet
Fault Detected
70
within the zone for any disturbance to be identified as a fault. Here, 20
impedance estimation points corresponds to 20 samples, which is equivalent to
a quarter of a cycle or 4.16 ms at 60 Hz. This is to ensure that the distance
algorithm does not mal-operate for small disturbances in the system, thus
providing robustness to the scheme. Further, this improves capability of the
scheme to provide correct discrimination between zones for near-boundary
faults as well (ex: 81% fault).
Validation of the Implemented Distance Protection
Scheme
The single line diagram (SLD) of the simulated power system section,
upon which the distance protection scheme is tested and validated is given
below in Figure 24.
Figure 24: SLD of the simulated power system using which the implemented distance
protection scheme is validated
System
Equivalent 2
Bus 2 Bus 1
VT CT
System
Equivalent 1
Load
71
The basic power system section simulated in RTDS is a 230 kV, 60 Hz,
200 km long overhead transmission line, connecting two AC networks. This
system is used to test and validate the SV acquisition and basic calculations
such as phasor extraction of the IED as well as the developed distance
protection algorithm. Figure 12 in Section 3.1, which illustrates the apparatus
used for testing the distance protection in detail, is repeated here in Figure 25
for the convenience of reference.
Figure 25: Schematic of the setup for testing the distance protection scheme
For this study, currents flowing out of the busbars are measured as
positive and the simulation time-step used is 50 μs. The transmission line is
GOOSE (trip) Protection IED
(on computer)
MU (GTNET1)
GTNET2
Circuit
Breaker IED
VT CT
Fault
Process
Bus
Transmission line
L
o
a
d
Sampled Values
RTDSTM
Bus 1
Switch
Switch
Bus 2
72
modelled using a frequency dependent transmission line model (frequency
dependent phase model) and the main parameters used are given in Appendix
B. Current transformers (CT) and voltage transformers (VT) are modeled using
standard library components in the RTDS. CT and VT ratios are 100 and 2000,
respectively. A dynamic load is connected at bus 2 and its active and reactive
power consumptions can be set as needed. The AC networks are modeled as
Thevenin’s equivalents with two 230 kV, 60 Hz sources. Impedances of the
network equivalents 1 and 2 are 176∠82° Ω and 200∠82° Ω, respectively.
Results Comparison for Impedance Estimations
Table 1 presents impedance estimation results, comparing the
impedance estimations taken from the IEC 61850 process bus compatible line
protection IED developed in the laboratory with the values estimated by a
commercial IED which takes current and voltage signals through the analog
outputs of the RTDS. These are also compared with impedance calculations
carried out in the RTDS, following similar steps as in the developed line
protection IED. The commercial IED used here is a line protection relay; L-
PRO 4000 manufactured by ERLphase Power Technologies Ltd. Moreover, the
positive and zero sequence impedances of the protected transmission line are
83.34∠86.43 Ω and 224.86∠72.30 Ω, respectively.
73
Table 1: Comparison of phase unit impedance calculations
Applied Faults
Primary Impedance seen by the phase unit AB
Laboratory
IED
L-PRO 4000 RTDS
Fault type Location
Magnitude
(Ω)
Angle
()
Magnitude
(Ω)
Angle
()
Magnitude
(Ω)
Angle
()
ABC
5% 4.67 79.57 4.60 88.00 4.74 80.86
50% 42.46 84.16 42.30 86.00 42.55 85.75
90% 77.36 85.06 76.40 86.00 77.44 86.28
AB
5% 4.75 79.91 4.50 84.00 4.70 80.82
50% 42.63 84.75 41.50 86.00 42.75 85.91
90% 77.20 85.59 76.30 86.00 77.65 86.27
The results presented in Table 1 and Table 2 show that the impedance
estimations carried out in the laboratory developed IED matches with those
taken from other sources to a large extent. Slight differences in the impedances
estimated by the commercial relay may be due to the differences in the inputs:
it takes voltages and currents through the analog outputs of RTDS simulator,
where some small deviation could be introduced based on the scaling factors
used in the digital to analog conversion block.
74
Table 2: Comparison of ground unit impedance calculations
Applied Faults
Primary Impedance seen by the ground unit AG
Laboratory
IED
L-PRO 4000 RTDS
Fault type Location
Magnitude
(Ω)
Angle
()
Magnitude
(Ω)
Angle
()
Magnitude
(Ω)
Angle
()
A-G
5% 4.71 73.12 4.60 76.00 4.74 75.81
50% 43.17 76.44 42.50 77.00 43.06 78.99
90% 77.77 78.12 76.40 79.00 77.22 80.09
AB-G
5% 4.32 74.61 4.50 84.00 4.43 76.89
50% 38.79 78.76 41.90 86.00 39.07 80.76
90% 69.85 80.29 76.20 86.00 71.01 81.98
ABC-G
5% 4.69 80.57 4.60 88.00 4.75 81.29
50% 42.16 85.81 41.60 87.00 42.78 86.20
90% 77.99 85.00 76.80 86.00 77.52 86.31
Estimated Impedance Trajectories
Some details of the operation of the implemented distance scheme are
examined to validate the correct operation. Figure 26 presents the convergence
of estimated impedance into the Mho circle for a zone 1 fault. The estimated
75
impedance point travels from the pre-fault value into the zone and settles on
the impedance line of the transmission line plotted in the R-X plane.
Figure 26: Trajectory of the estimated impedance of the line protection IED for a zone 1, AB
fault, plotted on a R-X plane; Zone 1 Mho characteristics and impedance line are shown
-30 -20 -10 0 10 20 30
0
10
20
30
40
50
60
70
80
R (primary Ohms)
jX (
prim
ary
Oh
ms)
-50 0 50 100 150 200 250-100
0
100
200
300
400
500
600
700
R (primary Ohms)
jX (
prim
ary
Oh
ms)
76
Fault Detection Time
Time taken for the detection of a fault is illustrated in Figure 27. Here,
Phase A current waveform, the estimated impedance scaled down by a factor
of 10 and the initiation of the trip signal (indicated as a binary status value of
the fault detection) is plotted.
Figure 27: Fault detection in the distance protection scheme for a 10 % fault
Form Figure 27, it is observed that the time elapsed from the reception
of the first SV packet after the fault to the fault detection is 73 samples, which
is under one cycle. This is well within the typically acceptable tolerance for a
distance protection algorithm. It is also observed that the time take by the
0 50 100 150 200 250 300 350
-10
-5
0
5
10
15
X: 139
Y: 1
Sample Number
Ph
ase
A C
urr
en
t Wa
vefo
rm (
Se
con
da
ry A
)E
stim
ate
d Im
pe
da
nce
*0.1
(S
eco
nd
ary
Oh
ms)
Fa
ult
De
tect
ion
X: 66
Y: 6.349
Current Waveform
Impedance Estimation
Fault Detection
77
algorithm for fault detection is slightly increasing with the distance to the
location of the fault. For a 90% fault, the number of samples taken by the
algorithm for fault detection is 92 (as indicated in Figure 28), whereas that for
a 10% fault is 73. This is a common characteristic of distance protection
algorithms, a one even commercial distance relays exhibit.
Figure 28: Fault detection in the distance protection scheme for a 90 % fault
Table 3 presents the time taken by the distance protection algorithm to
detect various types of faults in terms of number of samples. The two instances
that the algorithm fails to correctly identify the faults are high impedance
faults, where the estimated fault impedances lie outside the protected zone
settings of the relay.
0 50 100 150 200 250 300 350-6
-4
-2
0
2
4
6
8
10
12
14
X: 99
Y: 6.342
X: 191
Y: 1
Phase A
Curr
ent W
avefo
rm (
Secondary
A)
Estim
ate
d Im
pedance*0
.1 (
Secondary
Ohm
s)
Fault D
ete
ction
Sample Number
X: 99
Y: 0.5863
Current Waveform
Impedance Estimation
Fault Detection
78
Table 3: Fault detection time in terms of number of samples for distance protection algorithm
Applied Faults Time Taken
in number of samples (within
brackets the number of power
frequency cycles) Fault type Location Fault Impedance (Ω)
A-G
5% 0 58 (0.725)
50% 0 92 (1.15)
90% 0 92 (1.15)
A-G
5% 50 Doesn’t operate
50% 30 89 (1.11)
90% 30 Doesn’t operate
AB-G
5% 0 59 (0.728)
50% 0 82 (1.025)
90% 0 89 (1.11)
ABC-G
5% 0 66 (0.728)
50% 0 80 (1)
90% 0 115 (1.4375)
ABC
5% 0 66 (0.728)
50% 0 83 (1.0375)
90% 0 116 (1.45)
AB
5% 0 70 (0.875)
50% 0 86 (1.075)
90% 0 129 (1.6125)
79
Results for Series Compensated Transmission Lines
Distance protection faces some challenges in protecting series
compensated transmission lines. Due to voltage inversion, identification of
faults happening at different parts of the transmission line can become
difficult. The status of the bypass breaker, which may close during a fault,
could complicate the fault discrimination, if not known.
Figure 29 presents a simple example where a conventional distance
algorithm could malfunction due to the effect of the voltage inversion. It
illustrates the estimated impedance trajectories for a bus 1 fault, with and
without series compensation. Zone 1 Mho characteristic with a forward reach
of 80% is also shown. These results are obtained when the relay is fed with the
voltages measured on line side of the series capacitor.
It is obvious that with series compensation, the fault impedance is well-
within zone 1, erroneously indicating a line fault. For distance relays fed with
the line side voltage, certain reverse faults would appear as forward faults
when the series capacitor is in service.
80
Figure 29: Estimated impedance trajectories for a bus 1 fault (a) with, and (b) without series
compensation. Zone 1 Mho characteristic is also shown.
On the other hand, for relays fed with voltages measured on the bus side
of the series capacitor, certain zone 2 faults would appear as zone 1 faults,
causing the relay to overreach. This is shown in Figure 30 below. Zone 1 and
zone 2 have forward reach settings of 80% and 120 % of the protected
transmission line, respectively.
Due to these complications, it is advantageous to have more reliable
means of protecting transmission lines with series compensation. Analysis of
line current transients is one such method that has the potential to be
successful in this regard.
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2-1
-0.5
0
0.5
1
1.5
2
2.5
3
3.5
4
R (Secondary Ohms)
jX (
Se
co
nd
ary
Oh
ms)
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2-1
-0.5
0
0.5
1
1.5
2
2.5
3
3.5
4
R (Secondary Ohms)
jX (
Se
co
nd
ary
Oh
ms)
(a) (b)
81
Figure 30: Estimated impedance trajectories for a bus 1 fault (a) with, and (b) without series
compensation. Zones 1 and 2 Mho characteristic are also shown.
Chapter Summary
Chapter 4 presented the implementation details of the distance
protection scheme in the line protection IED. Detailed descriptions about
impedance estimation methods, development of impedance characteristics and
fault identification algorithms were also provided. Steps of the validation
process of the implemented scheme were then presented with a comparison of
results with commercial IEDs. Finally, results for protection of series
compensated lines were presented and associated protection challengers were
highlighted.
-3 -2 -1 0 1 2 3 4
-1
0
1
2
3
4
5
6
R (Secondary Ohms)
jX (
Se
co
nd
ary
Oh
ms)
-3 -2 -1 0 1 2 3 4-1
0
1
2
3
4
5
6
R (Secondary Ohms)
jX (
Se
co
nd
ary
Oh
ms)
(a) (b)
Chapter 5
Transients Based Protection Scheme
Typical sampling rates modern digital relays operate, range from 20 to
32 samples per cycle [33]. In contrast, the SV protocol, when implemented
according to IEC 61850-9-2 LE, has a significantly superior sampling rate and
hence, can be used to implement non-conventional protection algorithms such
as transients-based protection functions. It is observed that quite a number of
recently the published research work on SV based process bus has focused on
implementation and performance testing of protection algorithms [34], [35].
However, none of it seemed to have focused on transient protection, nor for
that matter, has exploited the high sampling frequencies associated with 9-2
LE based SVs. This section proposes a method of implementing a transient
based protection scheme that requires a high sampling rate as an application
of SV protocol in accordance with IEC 61850-9-2 LE.
83
Concept of the Proposed Method of Line Protection
This work concentrates on a series compensated overhead transmission
line, connecting two AC systems. At the terminating ends of the transmission
line exist two line protection relays, which receive instantaneous values of local
currents and voltages via Sampled Values (as per IEC 61850-9-2).
As explained earlier, occurrences such as faults, lightning strikes and
switching operations in the power system generate high frequency current
transients. Wave-front polarity of such a transient, observed at a particular
point in the power system, depends on the location of that point with respect
to the origin of the transient. The proposed method is based on the concept that
the locality of a fault (internal or external) can be identified by comparing the
polarities of the initial current transient seen at either end of the line [36]. The
application of the concept to a single circuit, transmission line is depicted in
Figure 31.
Figure 31: Origination of high frequency wave-fronts due to a fault
System
Equivalent 1 System
Equivalent 2
Bus 2 Bus 1
F2 F
1
B A
84
A fault in the system (such as F1 or F2) would generate current and
voltage travelling waves with wave-fronts of the same polarity, traveling away
from each other in opposite directions as shown. In the algorithm considered
in this thesis, current transients are considered. Here, points A and B at the
terminating ends of the line are considered as the two points of observation.
The directions of current measurements are as indicated. When a fault occurs
within the two points (i.e.: for an internal fault such as F1), it will be fed from
both ends. This results in a sudden increase (or decrease, depending on the
fault inception angle) in instantaneous current in the respective direction of
measurement at both points of observation, A and B. Thus, the wave-fronts
observed at either end are of the same polarity. In contrast, for an external
fault such as F2 behind bus-1, both points experience only the fault current
contribution coming from system 2, which is in the direction of observation at
point B, and opposite to the direction of observation at point A. Hence, the
observations at points A and B are of opposite polarities. This is true for any
external fault. The similarity or difference in the observed wave front polarities
at two line ends can, therefore, be utilized to distinguish between internal and
external faults of a transmission line (a more rigorous explanation is given in
Appendix D).
Nonetheless, it is important to understand that the polarity of the wave-
front observed at a particular point depends on the inception angle of the fault
as well. Moreover, several other factors including fault type, fault impedance,
85
line parameters as well as the sampling frequency and the signal processing
technique used for filtering may have a bearing on the exact signature of such
a high frequency transient.
As mentioned before, Wavelet Transform is a commonly used technique
for analyzing fast aperiodic electromagnetic transients. In this study, Discrete
Wavelet Transform (DWT) is used to decompose the observed currents into
frequency bands and extract the high frequency transients superimposed on
the power frequency currents. Instantaneous values of line currents are
converted into model components using constant Clarke transformation before
extracting the high frequency transients from them. This transformation
separates the ground mode component of the observed three phase currents
that generally travels at a slower velocity, and allows accurate detecting and
discriminating of faults of all different types by analyzing only the areal mode
components.
Implementation of the Transient Protection Scheme
5.2.1 Clarke Transformation of the Line Currents
As the first step of implementation, the instantaneous values of three
phase line currents subscribed by the relay as SVs are converted into model
components using the constant Clarke transformation matrix given in
equation (28).
86
(
𝐼0
𝐼𝛼
𝐼𝛽
) =1
3(
1 1 12 −1 −1
0 √3 −√3) (
𝐼𝑎
𝐼𝑏
𝐼𝑐
)
From the above transformation matrix, it is clear that all types of faults
can be analysed by only using α and β components (I.e. α component contain
transients for all types of fault expect those occurring between phases B and
C. And β component reflects transients for all types of faults except an A-G
fault).
5.2.2 Performing the Discrete Wavelet Transform
The discrete wavelet transform is performed using the multi-level filter
bank implementation as explained in Section 0. However, only the level 1
detail coefficients are of interest for this work, for those yield the transient
components in the highest available frequency band. Also due to limitation of
computing resources, only Iα is analysed in the implemented prototype relay.
In an actual relay, I is also needed to be processed in a similar manner for it
to enable detecting Phase-BC faults.
For the implementation of the DWT, “db4” (Daubechies 4) is chosen as
the mother wavelet, after carrying out studies with several wavelet families on
a trial and error basis (and also depending on the recommendations given in
[19], [20]). Wavelet function of the db4 mother wavelet is given in Figure 10.
The filter coefficients of the high pass filter, h[n], are given in the Appendix C.
87
Figure 32: Implementation of the DWT
Figure 33 presents the high frequency transients extracted from Iα using
the DWT. It can be observed very clearly that the emergence of this transient
coincides with the increase in the current due to the fault, hence the transient
is understood to have been caused by the fault itself. Moreover, from the
zoomed version of the transient, it can be seen that the wave-front polarity in
this particular instance is negative.
Figure 33: Extracted high frequency transient from Iα (for a sampling frequency 80 samples
per cycle)
Figure 34 shows the extracted high frequency transients from the same
signal, when the sampling frequency of 256 samples per cycle (15.36 kHz).
Iα/ I
Level 1 detail
coefficients Ia, Ib, Ic
h[n]
Clarke
transformation
0 50 100 150 200 250-10
0
10
20
Sample Number
Alp
ha c
om
ponent
(secondary
A)
0 50 100 150 200 250-0.02
-0.01
0
0.01
0.02
Sample Number
Level 1
deta
il coeffic
ients
0 10 20 30 40 50 60 70 80 90 100-0.02
-0.01
0
0.01
0.02
Sample Number
Level 1
deta
il coeffic
ients
(zoom
ed)
88
These results obviously provide more information than those for 80 samples
per cycle sampling frequency, nonetheless, do not make a qualitative difference
to the proposed algorithm.
Figure 34: Extracted high frequency transient from Iα (for a sampling frequency 256 samples
per cycle)
0 100 200 300 400 500 600 700-10
0
10
20
Sample Number
Alp
ha c
om
ponent
(secondary
A)
0 100 200 300 400 500 600 700-1
-0.5
0
0.5
1
Sample Number
Level 1
deta
il coeffic
ients
0 50 100 150-1
-0.5
0
0.5
1
Sample Number
Level 1
deta
il coeffic
ients
(zoom
ed)
89
Figure 35: Detail wavelet coefficients at the near end for the same fault with a sampling rate
of (a) 256 samples per cycle (b) 80 samples per cycle
Figure 35 presents a comparison of wavelet coefficients obtained at the
near end for the same fault during a single cycle at 60 Hz. Results for both the
sampling rates have the transient in them although their magnitudes and
frequencies are different. Further, it is clear that both the sampling rates
provide enough information (about the wave-front polarity) to be used in the
proposed protection algorithm.
Notice that the near end refers to Bus 1 and the far end refers to Bus 2
of the transmission line, as shown in Figure 31. Furthermore, all distances (for
instance, distance to the location of a fault) are measured from Bus 1.
0 50 100 150 200 250-1
-0.5
0
0.5
1
Sample Number
Level 1 d
eta
il
coeff
icie
nts
(a)
0 10 20 30 40 50 60 70 80-0.02
-0.01
0
0.01
0.02
Sample Number
Level 1 d
eta
il
coeff
icie
nts
(b)
90
Figure 36-(a) and (b) present the detail wavelet coefficients for the same
fault with fault inception angles 180° apart. There, the reversal of the wave-
front polarity is obvious. Further, (a), (c) and (d) of Figure 36 reveal the change
of the signature of the transient with the location of the fault. It is evident that
the observed transient is sustained for a longer period of time as the location
of the fault moves away from the point of observation. The observed wave-front
polarities, however, remain the same for faults at all three locations.
Figure 36: Detail wavelet coefficients at near end for (a) a 10% fault with fault inception
angle at 0°, (b) a 10% fault with fault inception angle at 180°, (c) a 50% fault with fault
inception angle at 0° and (d) a 90% fault with fault inception angle at 0°.
0 5 10 15 20 25 30 35 40-0.06
-0.04
-0.02
0
0.02
0.04
0.06
Sample Number
Le
ve
l 1
de
tail
co
effic
ien
ts
at th
e n
ea
r e
nd
(a)
0 5 10 15 20 25 30 35 40-0.06
-0.04
-0.02
0
0.02
0.04
0.06
Sample Number
Le
ve
l 1
de
tail
co
effic
ien
ts
at th
e n
ea
r e
nd
(b)
0 10 20 30 40 50 60 70 80-0.08
-0.06
-0.04
-0.02
0
0.02
0.04
0.06
Sample Number
Le
ve
l 1
de
tail
co
effic
ien
ts
at th
e n
ea
r e
nd
(c)
20 40 60 80 100 120-0.06
-0.04
-0.02
0
0.02
0.04
0.06
Sample Number
Le
ve
l 1
de
tail
co
effic
ien
ts
at th
e n
ea
r e
nd
(d)
91
It is further observed form Figure 37 that the values of the detail
wavelet coefficients following a fault are far greater than those appearing
during the steady state. Therefore, a simple threshold can be used to identify
the occurrence of a fault. In Figure 38, absolute values of the detail wavelet
coefficients are shown with a pre-set threshold.
Figure 37: Detail wavelet coefficients at either end for a 50% fault
Figure 38: Absolute values of the detail wavelet coefficients at either end for a 50% fault
0 10 20 30 40 50 60-0.06
-0.04
-0.02
0
0.02
0.04
0.06
X: 20
Y: -0.01889
Sample Number
Le
ve
l 1
de
tail
co
effic
ien
ts
at th
e n
ea
r e
nd
X: 16
Y: -6.06e-07
X: 21
Y: 0.05175
0 10 20 30 40 50 60-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
X: 16
Y: -1.12e-06
Sample Number
Le
ve
l 1
de
tail
co
effic
ien
ts
at th
e fa
r e
nd
X: 20
Y: -0.07339
X: 21
Y: 0.2459
20 30 40 50 60 700
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
X: 59
Y: 0.005
Sample Number
Le
ve
l 1
de
tail
co
effic
ien
tsa
t th
e n
ea
r e
nd
(a)
X: 32
Y: 0.03015
20 30 40 50 60 700
0.05
0.1
0.15
0.2
0.25
X: 56
Y: 0.005
Sample Number
Le
ve
l 1
de
tail
co
effic
ien
tsa
t th
e fa
r e
nd
(b)
X: 32
Y: 0.0994
92
The value of the first point above the threshold is also indicated in
Figure 38. The threshold is chosen empirically, based on the values observed
for the detail wavelet coefficients for a larger number of faults and other
disturbances. This is determined by adding some safety factor to the noise level
observed during normal operation. It should also be checked against the ability
of detecting high impedance faults (reasonably). There exists a compromise
between the algorithm’s capability to correctly identify the wave-front polarity,
and its susceptibility to noise. In other words, if the threshold is set too high,
the algorithm might miss the wave-front (especially, that of a high impedance
fault), and intern, if it’s too low, it can be trigged by noise.
5.2.3 Fault Identification
The fault identification method is as follows. The absolute value of the
extracted high frequency transient (detailed coefficients of DWT) is computed
and compared against a pre-set threshold value. When a coefficient value
passes the threshold then, the original sign of that value is saved as the
polarity of the wave-front. Once a fault is detected, a corresponding GOOSE
message will be send, depending on the observed polarity. The reception of this
GOOSE message would indicate the potential detection of a fault as well as the
wave-front polarity of the fault generated current. The process of generating
this GOOSE message is illustrated in Figure 39.
93
Figure 39: Sending of GOOSE messages corresponding to wave-front polarity
As shown in Figure 39, both positive and negative polarities have been
assigned to a GOOSE output variable each, which initiates the corresponding
GOOSE message once a fault detected. The line protection IED at the remoted
end subscribes the incoming message and compares the received wave-front
polarity with its own. The IED at near end similarly subscribes the
corresponding GOOSE messages sent from the remote end IED. The actions of
the two IEDs are determined by the two wave-front polarities as described in
Section 5.1.
Hybrid Protection Algorithm
From the earlier discussion on current transients in Section 5.1, it is
clear that certain non-faulty conditions such as switching operations can also
generate transients in line currents. This may cause a scheme that depends on
wave-front polarities as the sole method of fault identification to mal-operate.
In order to overcome this issue and also to increase the reliability and the
Wave-
front
polarity
Negative
Positive
GGIO1.ST.Ind1.stVal
IED at the far end
GGIO1.ST.Ind1.stVal
IED at the near end
GGIO2.ST.Ind1.stVal GGIO2.ST.Ind1.stVal
Fault detected
94
robustness of the scheme, a new hybrid protection algorithm is introduced by
combining the transient based protection algorithm with distance protection.
This combination enables the proposed method to effectively prevent non-
faulty transient producing events from being identified as faults. The proposed
hybrid line protection scheme is designed to operate similar to a transfer-trip
scheme as shown in Figure 40.
Figure 40: Illustration of the transfer-trip scheme
Upon detection of a disturbance, two protection IEDs (relays) R1 and R2
exchange the observed wave-front polarities with each other through GOOSE
messages, as shown in Figure 40. Then at each IED, the locally observed wave-
front polarity is compared with the remote end wave-front polarity received via
GOOSE messages. If the two wave-fronts are of the same polarity, transient
based line protection algorithm declares an internal fault. However, this
transient based protection logic is supervised by a local measurement based
distance element: a trip signal will be issued, only if the fault is confirmed by
the supervising distance elements with modified characteristics.
System
Equivalent 1 System
Equivalent 2
IEC 61850-90-5
inter-substation
communication
R2
Load
Bus 2 Bus 1
R1
95
A supervising distance element used in the hybrid algorithm is
implemented with a zone having Mho characteristics, but with modified reach
settings: a forward reach of 110% and a reverse reach of 10%, as shown in
Figure 41. These reach values are selected to ensure that the estimated
impedance would lie within the Mho circle for faults at all locations on the line,
with the series capacitor in and out of service, assuming that voltage
measurements are taken at line side of the series capacitor. The dashed line in
Figure 41 is the transmission line impedance. The combined algorithm for fault
detection is given below in Figure 42. According to the algorithm presented in
Figure 42, the fault is detected and discriminated (whether it is inside the
protected zone or not) by the transient based protection scheme, but trip signal
is issued only if a line fault is confirmed by the distance protection.
Figure 41: Modified Mho characteristics for the proposed protection method
-10%
X
R
110%
96
Figure 42: Combined algorithm of fault detection for the proposed hybrid protection scheme
Same polarity
Instantaneous currents
and voltages Instantaneous
currents
Impedance
estimation
Detailed
coefficients Next packet Next packet
GOOSE
YES
NO NO
YES
Clarke's Transform
and DWT
Decode
Mimic filter, phasor
extraction, and impedance
estimation
Internal
fault, TRIP
SV packets
Get wave-front
polarity
Send GOOSE
Remote wave-
front polarity
Local wave-front
polarity
External disturbance,
Do not act
Opposite
polarity
Above
threshold?
Inside
zone?
Same or
Opposite?
97
Validation of the Hybrid Protection Scheme
The single line diagram (SLD) of the simulated power system section,
upon which the transient protection scheme is tested and validated, is given
below in Figure 43. The locations of the voltage and current measurements are
also indicated.
Figure 43: SLD of the simulated power system using which the implemented transient
protection scheme is validated
To explore the full potential of the transients-based hybrid protection
scheme, series compensation is introduced at the near end of the transmission
line with a percentage compensation of 40 %. Furthermore, the series
capacitors are modelled using the thyristor protected models with MOVs and
bypass switches (important parameters are given in Appendix B). All other
system parameters are similar to those used for the validation of the distance
protection scheme.
The function of the relay at the far end of the line, R2, is also
implemented inside the RTDS simulation case. This is carried out employing
System
Equivalent
2
System
Equivalent 1
Load
Bus 2 Bus 1
Series
capacitor
CT VT
98
standard library components and using similar procedures and algorithms as
in the relay prototype implemented on desktop computer. This too detects
disturbances in the system and records the respective wave-front polarities as
seen at its location. Therefore, the two relays considered for this scheme are
the IED (R1 or IED1) and its virtual equivalent implemented in the RTDS case
(R2 or IED2). Figure 44 below illustrates the corresponding schematic of the
test apparatus.
Figure 44: Schematic of the developed system for the hybrid protection scheme
Notice that, only one link of the transfer trip scheme (communication
from IED1 to IED2) is implemented in the validation process. Since the
operation of the protection system is not real-time (as explained in Section 3.4),
Bus 1 Bus 2
System 2
Load
Series
capacitor System 1
Circuit Breaker
at far end
MU
(GTNET1)
CT VT
RTDS
Trip
IED2
CT VT
GTNET2
GOOSE
(Wave-front polarity)
Process Bus
Line Protection Relay on
desktop computer; IED1
SV
Switch
Switch
Virtual Line
Protection Relay
implemented in
RTDS
99
implementation of the other link does not yield a meaningful result. As shown
in Figure 44, upon detection of a fault by the line protection IED (IED1), the
wave-front polarity observed is sent via a GOOSE message (indicated in red)
to IED2 at the far end of the line, which is implemented inside the RTDS. Then
the IED2 compares it with the locally observed wave-front polarity and acts
accordingly. In case of an internal fault, it would issue a trip signal to the
circuit breaker at the far end to operate. Inter-substation communication
between IEDs 1 and 2 are envisioned to be carried out using routable GOOSE
messages as per IEC 61850-90-5. For this work, however, the said
communication is accomplished by regular GOOSE messages, as both IEDs are
located in the laboratory and connected to the same LAN. Although the second
link is not implemented, the sending of the GOOSE message with the wave-
front polarity observed by IED2 is indicated in the green line.
The graphs in Figure 45 present the three phase currents, voltages,
model current component, detail WTCs (Wavelet Transform Coefficients),
trajectories of the estimated impedance and the fault detection at either end
during the fault. According to the WTCs computed at the two current
measurement points, the fault is an internal fault. The estimated impedances
at both ends converge and settle inside the special Mho zones defined for the
respective relays, confirming the occurrence of a fault. The dotted blues lines
the bottom most plots indicates the times when transient based algorithms
detect the fault and initiate respective GOOSE messages with the polarity of
100
the wave front. The red solid lines show the times when the trip signals are
initiated, after the confirmation by the impedance units.
Figure 45: Three phase currents, voltages, model current component (Iα), detail WTCs,
trajectories of the estimated impedance and the fault detection at either end for a 50%, ABC
fault.
0.04 0.06 0.08 0.1-2
0
2
time (s)
I fault (
kA
)Near End
0.04 0.06 0.08 0.1-2
0
2
time (s)
Far End
0.04 0.06 0.08 0.1-200
0
200
time (s)
Vbus (
kV
)
0.04 0.06 0.08 0.1-200
0
200
time (s)
0.04 0.06 0.08 0.1-10
0
10
time (s)
I alp
ha (
A)
0.04 0.06 0.08 0.1-5
0
5
10
time (s)
10 20 30 40-0.01
0
0.01
deta
il W
CT
s
10 20 30 40-0.05
0
0.05
-3 -2 -1 0 1 2 3
0
2
4
R (Ohms)
jX (
Ohm
s)
-3 -2 -1 0 1 2 3
0
2
4
R (Ohms)
0.04 0.06 0.08 0.1-0.5
0
0.5
1
1.5
time (s)
Fault
Dete
ction
0.04 0.06 0.08 0.1-0.5
0
0.5
1
1.5
time (s)
101
When the line is series compensated, there exist two options of voltage
measurement at the near end (at bus 1) as mentioned in Section 2.2.2. They
are the bus side voltage VB and the line side voltage VL, as given in Figure 46.
Typically, VL is used in series compensated line protection, therefore, all
algorithms and settings in this application are devolved based on the
assumption that the relay at the near end is fed with VL.
Figure 46: Voltage measurements at the bus side and the line side
Figure 47 presents the results for a bus 1 fault with a 40% of series
compensation. A bus 1 fault is external to the two points of observation,
therefore, the WTC are of opposite polarities. The algorithm corrected
identifies this as an external fault although the estimated impedances are well
within the zone.
System Equivalent 1
Bus 1
Series
capacitor
VT CT VT
VB VL
102
Figure 47: Trajectories of the estimated impedance and detail wavelet coefficients at either
end for a bus 1, ABC fault, with the effect of voltage inversion
For certain reverse faults, the impedance seen by a distance relay at bus
1 would appear inside the forward protected zone. This is due to the effect of
voltage inversion across the series capacitor. A conventional distance relay
with a zone-1 forward reach characteristic of 80%, will misidentify such faults
as internal faults, even though they are actually external faults.
0 10 20 30 40
-0.5
-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
0.4
0.5
Sample Number
Deta
il C
oeff
icie
nts
0 10 20 30 40
-0.25
-0.2
-0.15
-0.1
-0.05
0
0.05
0.1
0.15
0.2
0.25
Sample Number
Deta
il C
oeff
icie
nts
-3 -2 -1 0 1 2 3-1
0
1
2
3
4
5
R (Ohms)
jX (
Ohm
s)
Near End
-3 -2 -1 0 1 2 3-1
0
1
2
3
4
5
R (Ohms)
jX (
Ohm
s)
Far End
103
Figure 48 and Figure 49 present similar results when the relay is fed
with the bus side voltage, VB, for a forward fault at 5% of the line.
Figure 48: Trajectory of the estimated impedance for a 5% three phase fault with series
capacitor in service, (with the effect of voltage inversion)
Figure 49: Detail wavelet coefficients at either end for a 5% three phase fault with series
capacitor in service, (with the effect of voltage inversion)
When the voltage measurements are taken from the bus side of the
capacitor, the reverse reach of the Mho circle must be increased as the
impedance seen by the relay can now be negative. Here, a reverse reach of 50%
is used. Now, for a bus 1 fault, although the estimated impedance lies within
-5 -4 -3 -2 -1 0 1 2 3 4 5
-3
-2
-1
0
1
2
3
4
5
R (Secondary Ohms)
X (
Secondary
Ohm
s)
0 5 10 15 20 25 30
-4
-2
0
2
4
x 10-3
Sample Number
Level 1 d
eta
il coeff
icie
nts
at
the n
ear
end
0 5 10 15 20 25 30-0.03
-0.02
-0.01
0
0.01
0.02
0.03
Sample Number
Level 1 d
eta
il coeff
icie
nts
at
the f
ar
end
104
the negative zone, the wave-front polarities are still of the same polarity.
Hence, it is correctly identified as an internal fault.
The results presented above demonstrate the basic operation of the
algorithm. That is, a fault is identified as internal only if the wave-front
polarities are of the same polarity as well as the impedance seen is within the
predefined zone.
The following (plotted in Figure 50, Figure 51, Figure 52 and Figure 53)
are the wave-front polarities recorded at either end at several chosen locations
of the transmission line for different scenarios.
Figure 50: Detail wavelet coefficients at either end for a 10% AB fault
20 25 30 35 40 45 50 55 60 65 70-0.06
-0.04
-0.02
0
0.02
0.04
0.06
Sample Number
Le
ve
l 1
de
tail
co
effic
ien
ts
at th
e n
ea
r e
nd
20 25 30 35 40 45 50 55 60 65 70-0.3
-0.2
-0.1
0
0.1
0.2
Sample Number
Le
ve
l 1
de
tail
co
effic
ien
ts
at th
efa
r e
nd
105
Figure 51: Detail wavelet coefficients at either end for a 50% AB fault
Figure 52: Detail wavelet coefficients at either end for a 90% AB fault
30 40 50 60 70 80 90
-0.05
0
0.05
Sample Number
Le
ve
l 1
de
tail
co
effic
ien
tsa
t th
e n
ea
r e
nd
30 40 50 60 70 80 90-0.2
-0.1
0
0.1
0.2
0.3
Sample Number
Le
ve
l 1
de
tail
co
effic
ien
tsa
t th
e fa
r e
nd
0 10 20 30 40 50 60-0.1
-0.05
0
0.05
0.1
Sample Number
Le
ve
l 1
de
tail
co
effic
ien
tsa
t th
e n
ea
r e
nd
0 10 20 30 40 50 60-0.2
-0.1
0
0.1
0.2
Sample Number
Le
ve
l 1
de
tail
co
effic
ien
tsa
t th
e fa
r e
nd
106
Figure 53: Detail wavelet coefficients at either end for an external fault (a bus 1 fault)
Above Figure 50, Figure 51 and Figure 52 show that the wave-front
polarities observed at each end are of the same polarity for internal faults at
different locations. On the other hand, for a bus 1 fault, which is external to
the two points of observation, the observed wave-fronts are of opposite
polarities as seen in Figure 53. These results, therefore, verify the theory
formulated earlier regarding the fault identification using wave-front
polarities. It is further noticed that the magnitude of the wavelet coefficients
and the time of arrival of the wave-front understandably depend on location of
the fault. For instance, for a 90% fault, the initial wave-front reaches the far
end earlier than its counterpart reaches the near end, and the magnitude of
the former is larger than that of the latter. These magnitudes, however, are
also affected by other factors such as the type of the fault, pre-fault load current
0 10 20 30 40 50 60-0.4
-0.2
0
0.2
0.4
Sample Number
Le
ve
l 1
de
tail
co
effic
ien
ts
at th
e n
ea
r e
nd
0 10 20 30 40 50 60-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
Sample Number
Le
ve
l 1
de
tail
co
effic
ien
ts
at th
e fa
r e
nd
107
magnitudes, source and fault impedances and also the inception angle.
Although this is not detrimental to the function of the proposed hybrid
algorithm, the threshold value against which the WTCs are compared to detect
faults may need adjustments depending on the power system.
One of the most challenging tasks of transmission line protection is to
effectively protect lines with series compensation. Traditional distance
protection schemes tend to mal-function due to current or voltage reversals
caused by the series capacitor. Another difficulty faced in distance protection
is the detection of high impedance faults. Schemes with Mho characteristics,
in particular, provide very little resistive coverage for downstream faults,
while quadrilateral characteristics perform better in this regard. Also, for the
faults located close to the far end (beyond 80%), the trip signal is delayed with
traditional distance protection (without a transfer trip scheme) as they are
detected in Zone 2.
The proposed method of line protection can be adopted to address both
the issues mentioned above. While the proposed algorithm can effectively
protect a series compensated line with existing settings, its protected zone
(modified distance protection characteristic in Figure 41) is required to be
redefined for it to cover high impedance faults as well. This is required in order
to provide the necessary resistive coverage. In Figure 54, the detail wavelet
coefficients at either end for a 5% fault with 40% series compensation are
shown. And similarly in Figure 55, those for a 95% fault with a fault impedance
108
of 50 Ω (without series compensation) are shown. The wave-front polarities in
both occasions correctly indicate an internal fault.
Figure 54: Detail wavelet coefficients at either end for a 5% AB fault with 40% series
compensation
Figure 55: Detail wavelet coefficients at either end for a 95% AG fault with a 50 Ω fault
impedance
0 10 20 30 40 50 60
-0.02
-0.01
0
0.01
0.02
Sample Number
Le
ve
l 1 d
eta
il co
effic
ien
tsa
t th
e n
ea
r e
nd
0 10 20 30 40 50 60-0.4
-0.2
0
0.2
0.4
Sample Number
Le
ve
l 1 d
eta
il co
effic
ien
tsa
t th
e fa
r e
nd
20 30 40 50 60 70 80 90 100-4
-2
0
2
4
x 10-3
Sample Number
Le
ve
l 1 d
eta
il co
effic
ien
tsa
t th
e n
ea
r e
nd
20 30 40 50 60 70 80 90 100-0.02
-0.01
0
0.01
0.02
Sample Number
Le
ve
l 1 d
eta
il co
effic
ien
tsa
t th
e fa
r e
nd
109
Since the proposed hybrid protection scheme covers full length of the
line, there is no intentional delay for any fault within the protected zone.
However, the operation of transient based protection requires communication
of wave-front polarities. However, this communication does not cause
additional delay. Detection of fault and evaluation of wave-front polarities is
very fast (less than 1 ms) and therefore, GOOSE messages from the respective
remote ends can arrive at each IED well before the fault is confirmed by its
distance element, which takes about one cycle (16.67 ms). This is clearly
evident from Figure 56, which illustrates the time line of fault detection
process. In this example, the time taken by the distance element is 86 samples,
which is equivalent to 1.075 of a cycle at 60 Hz, while the detection by the
transient based protection algorithm is almost immediate.
Figure 56: Fault detection in the proposed protection scheme for a solid AB fault, 5% away
from bus-1.
0 50 100 150 200 250
-10
-5
0
5
10
15
X: 85
Y: 6.742
Sample Number
X: 87
Y: 1
X: 171
Y: 1
Phase A
Curr
ent
(Secondary
A)
0.1
*Im
pedance E
stim
ation (
Secondary
Ohm
s)
Fault D
ete
ction (
both
tra
nsie
nt
and d
ista
nce p
rote
ction)
Current w aveform
Impedance estimation
Fault Detection (Transient Protection)
Fault Detection (Distance Protection)
110
Performance Evaluation of the Hybrid Protection
Scheme
Performance of the proposed transient protection scheme is tested under
a variety of faults and the results are tabulated below in Table 4.
Table 4: Performance of the proposed protection scheme under different fault conditions
Applied Faults
Fault Detection Fault
type
Location
Fault
Impedance (Ω)
Series
compensation
AB
5% 0 Yes Yes
50% 0 Yes Yes
95% 0 No Yes
ABC
5% 0 Yes Yes
50% 0 No Yes
95% 0 No Yes
A-G
5% 0 Yes Yes
50% 0 No Yes
95% 50 No No
AB-G
5% 0 Yes Yes
50% 30 No Yes
95% 50 No No
ABC-G
5% 0 Yes Yes
50% 0 Yes Yes
95% 0 Yes Yes
ABC Bus 1 0 No No
111
With the current distance characteristic, the scheme doesn’t work for
high impedance faults occurring at far downstream locations, simply because
the impedance point lies outside the protected zone. This hybrid protection
scheme is also tested for a variety of percentage line compensations and for
different source to impedance (SIR) ratios.
5.5.1 Behaviour for Non-faulty Conditions
The transients based protection scheme is also tested for some non-
faulty conditions such as circuit breaker operations. Figure 57 shows the wave-
front polarities observed for a closing operation of the circuit breaker at the
near end.
Figure 57: Detail wavelet coefficients at either end for a closing operation of the circuit
breaker at the near end
20 30 40 50 60 70 80-0.2
-0.1
0
0.1
0.2
0.3
Sample Number
Le
ve
l 1
de
tail c
oe
ffic
ien
ts
at th
e n
ea
r e
nd
20 30 40 50 60 70 80-6
-4
-2
0
2
4x 10
-3
Sample Number
Le
ve
l 1
de
tail c
oe
ffic
ien
ts
at th
e fa
r e
nd
112
Notice that the two wave-fronts are of opposite polarities similar to an
external fault, despite the fact that it is an internal event (in this case).
Therefore, this would not even trigger the transients-based section of the
algorithm (impedance relay is obviously not triggered by this). Moreover, a
similar behaviour is observed for the operation of the bypass switch of the
series capacitor as shown in Figure 58.
Figure 58: Detail wavelet coefficients at either end for a closing operation of the bypass
switch of the series capacitor.
0 5 10 15 20 25 30 35 40 45 50-2
-1
0
1
2
3x 10
-3
Sample Number
Le
ve
l 1
de
tail c
oe
ffic
ien
ts
at th
e n
ea
r e
nd
0 5 10 15 20 25 30 35 40 45 50-5
0
5
10x 10
-3
Le
ve
l 1
de
tail c
oe
ffic
ien
ts
at th
e fa
r e
nd
Sample Number
113
Effect of the Bit Resolution of the A/D Converter
The data types used in IEC 61850-9-2 LE for the representation
instantaneous values of currents and voltages are “INT32” and “FLOAT32”.
Both the data types are in 32 bit format. The RTDS has the capability to
publish sampled valued with a resolution of 32 bits. However, the hardware
analog to digital (A/D) converters in real merging units do not usually possess
a 32 bit resolution. Typically, modern commercial merging units have a 16-20
bit resolution in their A/D converters [37], [38]. It is discovered that this bit
resolution is of the utmost importance to the proposed transients based
protection scheme. The analysis carried out regarding this showed that the
high frequency transients become indistinguishable due to added noise, if the
resolution of the currents measured are below a certain level. The following
calculations are carried out assuming a bit resolution of 24.
Rated primary current of the CT = 100 A (rms). The CT ratio used is
100, therefore, the rated secondary current, Isec is 1 A. Peak of the rated
secondary current, Isec, pk is 1√2 A = 1.41421 A. Assuming an accuracy limit of
20 for the CT, the maximum current that can be accurately measured by the
CT is 20.Isec, pk.
20 = 24.322 → 5 bits
114
Saving these 5 bits and 1 bit for the sign, would leave 18 bits to represent
the regular secondary current. Therefore, the resolution of the secondary
current provided by the MU is,
1.41421
218= 5.3948 × 10−6A = 5.3948 μA
Current resolutions calculated for 18 bit, 16 bit and 12 bit A/D
converters, following the same procedure, are 0.3453 mA, 1.3811 mA and
22.097 mA, respectively. The wavelet coefficients observed after setting above
current resolutions are given in Figure 59 below.
115
Figure 59: Detail wavelet coefficients at the near end for a 50% fault, with an assumed A/D
converter bit resolution of (a) 24 bits, (b) 18 bits (c) 16 bits and (d) 12 bits
20 30 40 50 60 70 80 90 100
-0.05
0
0.05
(a)
20 30 40 50 60 70 80 90 100-0.1
-0.05
0
0.05
0.1
Level 1 d
eta
il c
oeffic
ients
at th
e n
ear
end
(b)
20 30 40 50 60 70 80 90 100-0.05
0
0.05
(c)
20 30 40 50 60 70 80 90 100
-0.1
-0.05
0
0.05
0.1
Sample Number
(d)
116
From the above results, it is clear that it is impossible to perform an
analysis of transients of the type discussed in this thesis, with a bit resolution
of 12. Results obtained for a bit resolution of 16 are also unworkable for
identifying the wave-front polarity, although one can recognize the presence of
the transient. Results for both bit resolutions 18 and 24 presents distinct
transients and can be effectively used for the purpose of this study. It should
be further noticed that the above calculations have not accounted for the noise
generated by the electronic devices and other means. Generally, 2-3 bits are
typically reserved as a safety margin for errors caused by the system noise.
From the above discussion, it can be recommended that the A/D converter of
the MU should have a bit resolution of 20 or above for SVs (published by that
MU) to be used in a transient based protection scheme of this nature. (This
requirement is predominantly regarding the current measurement).
Figure 60: Modelling the effect of a 24 bit A to D converter
SVs CT secondary
currents
24 bit A to D
converter
(Virtual)
Anti-aliasing
filter
SV
Component
Merging Unit
modelled in
the RTDS
Round-off to 6
µA
117
All the studies regarding the transient based scheme of this thesis are
carried out assuming a bit resolution of 24. To model the effect of a 24 bit A/D
converter, the output values of the anti-aliasing filters (as in Figure 17) are
rounded off to the nearest 6 µA (from 5.3948 µA).
With the modelling the effect of a 24 bit A/D converter, now the
arrangement shown in Figure 60 implemented inside the RTDS, emulates the
function of an actual merging unit with a substantial accuracy.
Chapter Summary
In Chapter 5, the steps followed in developing the transients based
hybrid protection scheme were put forth. First the basic concept was described
followed by the details explaining its implementation in an actual power
system. This scheme too was validated using the RTDS and a number of results
were presented in this chapter. Finally, modelling the effect of the bit
resolution of an actual merging unit (MU) was analysed and important
recommendations were provided.
Chapter 6
Conclusions
Contributions and Conclusions
In this thesis, implementation of an IEC 61850-9-2 LE, sampled values
compatible line protection IED is presented. A distance protection algorithm
and a transient based hybrid protection algorithm are successfully
implemented on this IED, both of which are tested through hardware-in-the-
loop simulations performed using a RTDS simulator. The main contributions
made during the work done in completing this thesis and pertinent conclusions
are listed below.
An IEC 61850-9-2 LE sampled values compatible IED prototype was
developed on a Linux based desktop computer. This prototype IED
could successfully subscribe to sampled values published by a source
with the correct configurations, decode the information, and pass the
decoded sampled values to signal processing algorithms for further
processing.
119
An experimental setup was developed to test the process bus
compatible IEDs by augmenting a test setup previously developed to
test the station bus capability of IEC 61850 compatible IEDs. This
development resulted in a fully IEC 61850 compliant laboratory setup.
Two simultaneous signal processing algorithms, a Discrete Fourier
Transform based phasor extraction algorithm and a Discrete Wavelet
Transform based transient extraction algorithm, were successfully
implemented on the prototype IED. Accuracy of signal processing was
confirmed by extensive tests.
A distance protection algorithm was implemented using the phasors
extracted from IEC 61850-9-2 LE sampled values. Both Mho and
quadrilateral characteristics were implemented. Operation of the
distance protection was extensively tested and verified through
hardware-in-the-loop simulations.
A transient based protection algorithm was implemented on the
prototype IED using IEC 61850-9-2 LE sampled values as inputs. The
fault identification in this algorithm is based on the comparison of
polarities of the fault generated travelling wave-fronts measured at
two line ends. This algorithm was combined with distance protection
to implement a reliable hybrid line protection scheme applicable for
protection of series compensated transmission lines.
120
This hybrid line protection scheme based on IEC 61850-9-2 LE
sampled values was extensively tested and verified using the real-time
digital simulator. The performance of the transient based algorithm
and its sensitivity to various factors was investigated through
experiments.
Based on the studies conducted, it can be concluded that
When IEC 61850-9-2 sampled values are used as inputs, generic
hardware such as the desktop PC used in this research can be used to
implement process bus based IEDs. The prototype IED on desktop is
a convenient platform for testing novel protection algorithms.
The hybrid protection scheme proposed in this research can effectively
protect series compensated transmission lines from all types of faults
without being affected by the voltage inversion due to series capacitor.
The sampled values as per IEC 61850-9-2 LE contain enough
information for them to be used in the class of transient based
protection algorithms implemented in this thesis.
When used for transient based protection algorithms, the analog to
digital converter of the merging need to have a higher bit resolution.
For the algorithm implemented in this research, the required
minimum analog to digital conversion resolution was 20 bits.
121
The fully IEC 61850 compliant laboratory setup developed for the
purpose of this research can be used for further research on the IEC
61850 protection algorithms as well as for teaching purposes.
122
Future Work
The followings can be viewed as future extensions on the work presented
in this thesis.
Although the implemented prototype IED is adequate for verifying the
protection algorithms, there was a time delay in acquisition of IEC
61850-90-2 sampled values (as explained in Section 3.4). The cause for
this time delay could not be pinpointed, but it could be due to the
network card used or due to the fact that the developed application is
running on a regular operating system not optimized for real-time
operation. Rectifying this problem is needed for achieving improved
time performance and make the prototype more realistic.
The protection of series compensated double circuit transmission lines
is also an existing challenge in power system protection. The transient
based protection algorithm can be further improved to accomplish this
as well. One way of achieving this would be to incorporate a current
comparison of the two circuits to the existing algorithm.
Other protection schemes such as overcurrent protection can be added
to the line protection IED. Furthermore, with some modifications to
the existing libraries, a differential protection IED can also be
implemented by subscribing two sampled values streams.
123
Currents and voltages of the RTDS can be taken out as analog outputs,
and then amplified and fed into the inputs of a real merging unit in
the laboratory. This would provide the opportunity to use sampled
values from an actual merging unit in the implemented IED and to
investigate the issues accompanied with it, which might not be present
in the sampled values published by the RTDS.
References
[1] ABB Review (Aug. 2010), Special Report IEC 61850, [Online] Available:
https://library.e.abb.com/public/a56430e1e7c06fdfc12577a00043ab8b/3B
SE063756_en_ABB_Review_Special_Report_IEC_61850.pdf.
[2] K.P. Brand, C. Brunner and W. Wimmer, “Substation Automation
Handbook” UAC 2003, ISBN 3-85759-951-5, 2003
[3] E. Chikuni, “Power System and Substation Automation,” Cape Peninsula
University of Technology, South Africa, [Online]
Available:https://scadahacker.com/library/Documents/ICS_Basics/InTec
h%20-%20Power%20System%20and%20Substation%20Automation.pdf
[4] R.E. Mackiewicz, "Overview of IEC 61850 and Benefits," in Power
Systems Conference and Exposition, 2006. PSCE '06. 2006 IEEE PES,
vol., no., pp.623,630.
[5] IEC 61850. (2015, January 14). In Wikipedia, The Free Encyclopedia.
Retrieved on February 25th, 2015, from
http://en.wikipedia.org/w/index.php?title=IEC_61850&oldid=642524536.
[6] IEC standard for Communication Networks and Systems in Substations,
Part 5: Communication Requirements for Functions and Device Models,
IEC 61850-5, First Edition, (Jul. 2003).
125
[7] IEC standard for Communication networks and systems for power utility
automation, Part 8-1: Specific communication service mapping (SCSM) –
Mappings to MMS (ISO 9506-1 and ISO 9506-2) and to ISO/IEC 8802-3,
IEC 61850-8-1, Second Edition, (Jun. 2011).
[8] IEC standard for Communication networks and systems for power utility
automation, Part 9-2: Specific communication service mapping (SCSM) –
Sampled values over ISO/IEC 8802-3, IEC 61850-9-2, Second Edition,
(Sep. 2011).
[9] B.A.H. Wickremasuriya, K.K.M.S. Kariyawasam and A.D. Rajapakse “A
Laboratory Setup for Teaching IEC 61850 based Substation Automation,”
2015 CIGRÉ Canada Conference, in press.
[10] D. McGinn, M. G. Adamiak, M. Goraj and J. Cardenas, "Reducing
conventional copper signaling in high voltage substations with IEC 61850
process bus system," in PowerTech, 2009 IEEE Bucharest, vol., no., pp.1-
8, June 28 2009-July 2 2009.
[11] IEC standard for Communication Networks and Systems in Substations,
Part 7-1: Basic communication structure for substation and feeder
equipment - Principles and models, IEC 61850-7-1, First Edition, (Jul.
2003).
[12] IEC standard for Communication networks and systems for power utility
automation – Part 6: Configuration description language for
126
communication in electrical substations related to IEDs, IEC 61850-6,
Second Edition, (Dec. 2009).
[13] C. Fernandes, S. Borkar and J. Gohil, “Testing of Goose Protocol of
IEC61850 Standard in Protection IED,” International Journal of
Computer Applications (0975 – 8887) Volume 93 – No 16, May 2014.
[14] UCA International Users Group, "Implementation guideline for digital
interface to instrument transformers using IEC 61850,” R2-1, UCA Int.
Users Group, Raleigh, NC, USA, 2004.
[15] L. F. Santos, “Substation Automation Process Bus System design and
experiences with IEC 61850-9-2 and NCITs”, ABB Group,
[Online]Available:http://www02.abb.com/global/gad/gad02465.nsf/59bdd
cf5e4536d06c125778c0028fa3c/0bccb6b99be39eda83257a8b0056716c/$FI
LE/Process+Bus.pdf.
[16] IEC standard for Communication networks and systems for power utility
automation - Part 90-5: Use of IEC 61850 to transmit synchrophasor
information according to IEEE C37.118, First Edition (May 2012).
[17] H. Falk, SISCO, USA, “IEC 61850-90-5 - an Overview”, PAC World,
December 2012 Issue, Retrieved on December 30th, 2015, from
https://www.pacw.org/issue/december_2012_issue/iec_61850905_an_over
view/iec_61850905_an_overview/complete_article/1.html.
127
[18] J. M. Gers and E. J. Holmes, “Protection of Electricity Distribution
Networks” (2nd Edition), Institution of Engineering and Technology,
2004, ISBN: 978-0-86341-357-5.
[19] E. Bakie, C. Westhoff, N. Fischer and J. Bell, “Voltage and Current
Inversion Challenges When Protecting Series-Compensated Lines – A
Case Study,” in 42nd Annual Western Protective Relay Conference
Spokane, Washington, October 20–22, 2015.
[20] D.C. Robertson, O.I. Camps, J.S. Mayer and W. B. Gish, "Wavelets and
electromagnetic power system transients," IEEE Transactions on Power
Del., vol.11, no.2, pp.1050-1058, Apr 1996.
[21] V. Giurgiutiu and L. Yu, “Comparison of Short-time Fourier Transform
and Wavelet Transform of Transient and Tone Burst Wave Propagation
Signals For Structural Health Monitoring,” 4th International Workshop
on Structural Health Monitoring, Stanford University, Stanford, CA,
September, 2003.
[22] N. Mehala and R. Dahiya, “A Comparative Study of FFT, STFT and
Wavelet Techniques for Induction Machine Fault Diagnostic Analysis,”
7th WSEAS int. conf. on computational intelligence, man-machine
systems and cybernetics (CIMMACS '08)
[23] N. Perera and A. D. Rajapakse, “Series-Compensated Double-Circuit
Transmission Line Protection Using Directions of Current Transients”,
128
IEEE Transactions on Power Del., vol. 28, no. 3, pp.1566- 1575 , July
2013.
[24] N. Perera, A. D. Rajapakse, and T. Buchholzer, “Isolation of faults in
distribution networks with distributed generators,” IEEE Transactions
on Power Del., vol. 23, no. 4, pp. 2347–2355, Oct. 2008.
[25] N. Perera, A. D. Rajapakse, and D. Muthumuni, “Wavelet based transient
directional method for busbar protection,” presented at the Int. Conf.
Power Systems Transients (IPST), Delft, the Netherlands, Jun. 14–17,
2011.
[26] M. Misiti and Y. Misiti, Georges Oppenheim and Jean-Michel Poggi,
“Wavelet Toolbox User’s Guide for use with Matlab (Version 1)”, The
MathWorks, Inc., March 1996.
[27] Wavelet Properties Browser, In WAVELET BROWSER BY
PYWAVELETS, Retrieved on December 30th, 2015, from
http://wavelets.pybytes.com/.
[28] RTDS Technologies Inc. “RTDS Hardware Manual - GTNET Network
Interface Card,” January 2014 rev 03.
[29] Systems Integration Specialists Company, Inc. (SISCO), “MMS Lite, IEC
61850 for Embedded Systems,” Retrieved on April 06th, 2016, from
http://www.sisconet.com/wp-content/uploads/2015/06/MktLit_MMS-
Lite_062015.pdf.
129
[30] G. Benmouyal, "Removal of DC-offset in current waveforms using digital
mimic filtering," IEEE Transactions on Power Del., vol.10, no.2, pp.621-
630, Apr 1995.
[31] M. Paolone, Synchrophasor fundamentals: from computation to
implementation, Tutorial, IEEE PES GM, Vancouver, July 21-25, 2013.
[32] K.P. Brand, C. Brunner and W. Wimmer, “Design of IEC 61850 based
Substation Automation Systems according to Customer Requirements,”
in CIGRE Plenary Meeting, Paris, 2004, Session of SC B5, Paper B5-103.
[33] M. Kezunovic, J. Ren and S. Lotfifard, “Design, Modeling and Evaluation
of Protective Relays for Power Systems” Edition 1, Springer International
Publishing, 2016.
[34] M.G. Kanabar, T.S. Sidhu and M.R.D. Zadeh, "Laboratory Investigation
of IEC 61850-9-2-Based Busbar and Distance Relaying With Corrective
Measure for Sampled Value Loss/Delay," IEEE Transactions on Power
Del., vol.26, no.4, pp.2587-2595, Oct. 2011.
[35] D.M.E Ingram, P Schaub, R.R. Taylor and D.A. Campbell, "System-Level
Tests of Transformer Differential Protection Using an IEC 61850 Process
Bus," IEEE Transactions on Power Del., vol.29, no.3, pp.1382-1389, June
2014.
[36] N. Perera, K. Narendra and D. Fedirchuk, “Power system fault zone
detection”, U.S. Patent 20130015878 A1, Jan 17, 2013.
130
[37] PCS-221 Merging Unit, Datasheet, NR Electric Co., Ltd., Retrieved on
December 30th, 2015, from http://www.nrec.com/en/product/PCS-
221.html.
[38] Analog Merging Unit, Datasheet, VIZIMAX Inc., Retrieved on November
11th, 2015, from http://www.vizimax.com/knowledge-center/85-
english/mergingunit/266-datasheets.
Appendix A
I. IEC 61850 Service Modules
In section I of Appendix A, some of the important information regarding
the software libraries developed by Kalkitech Pvt. Ltd are provided.
Information given below are taken from the user manual provided (University
Of Manitoba, IEC61850 Library APIs. User Manual, V 0.4).
INITIALIZATION MODULE
List of APIs
Set_Log_Level
Initialize_61850_Library
Close_61850_Library
Configuration of Parameters
Definition of the structure
typedef struct
char startUp_File[K_STRING_SIZE];
char scl_conf_file[K_STRING_SIZE];
char ied_name[K_STRING_SIZE];
char goose_network_ID[K_STRING_SIZE];
char sv_network_ID[K_STRING_SIZE];
char accPoint_name[K_STRING_SIZE];
CONFIG;
132
With this structure, user can specify the configuration parameters.
startUp_File - The startup file to be used by the IEC 61850
library.
scl_conf_file - SCL Configuration file to be used by the IEC 61850
library.
ied_name - IED name as in SCL the configuration file.
goose_network_ID - Network card to be used by the Goose
publisher/subscriber.
sv_network_ID - Network card to be used by the SV
publisher/subscriber.
accPoint_name - Access point name as specified in SCL
configuration file.
SERVER MODULE
Server module handles all the IEC 61850 server functions. Before using
any API's in this module, the IEC 61850 server must be initialized by the
initialization module. Functions in this module include,
Server.
Report service.
Read and write variable values.
Register client writes by a user callback function.
Read the events occurred
133
NOTE: It will not be allowed to read and write variable values and details if
none of the functionality (goose publisher, goose subscriber, sampled value
publisher, sampled value publisher or server) are running.
List of APIs
Get_Total_Variable_Count
Get_Variable_Details
Start_Server
Start_Report_Service
Stop_Server
Stop_Report_Service
Read_Variable_Value
Write_Variable_Value
Register_Client_Writes
Get_Event_Count
Get_Event
GOOSE PUBLISHER MODULE
This module is responsible for the GOOSE publisher function. This
module publishes new GOOSE messages, when a dataset variable configured
in any goose control block changes. And this also handles the retransmission
of goose messages.
134
List of APIs
Start_Goose_Publisher
Stop_Goose_Publisher
Start_Goose_Retransmission
Stop_Goose_Retransmission
GOOSE SUBSCRIBER MODULE
List of APIs
Start_Goose_Subscriber
Stop_Goose_Subscriber
Register_Goose_Events
SAMPLED VALUE SUBSCRIBER MODULE
This module handles all the sampled value subscription related
activities. It discovers the configured sampled value streams from the network.
User shall select the sampled value stream to subscribe from the network.
Received sampled value packets will be stored, which allows user to analyze
them.
List of APIs
Start_SV_Subscriber
Stop_SV_Subscriber
Discover_SV_Streams
135
Register_SV_Stream
Unregister_SV_Stream
Get_Subscribed_SVPackets_Count
Get_SV_Packet
Definition of the Structure
typedef struct
short smp_count;
short smp_rate;
unsigned long seconds;
unsigned long micro;
unsigned long quality;
int smp_data_len;
char *smp_data;
SV_DETAILS;
Structure is used to exchange sampled value information between user
application and IEC 61850 library. IEC61850 library returns the subscribed
SV packet details using this structure.
smp_count - sample count.
smp_rate - sample rate.
Seconds - refresh time seconds.
micro - refersh time micro seconds.
quality - refresh time quality.
smp_data_len - sample data length.
smp_data - Pointer to the sample data buffer.
136
SAMPLED VALUE PUBLISHER MODULE
List of APIs
Start_SV_Publisher
Stop_SV_Publisher
Get_SV_Publisher_Streams
Fill_SV_Data
II. CID file of the RTDSTM for SV publication
Given below is the CID file of the RTDSTM used for the publication of sampled
values.
<?xml version="1.0" encoding="UTF-8"?>
<SCL xmlns="http://www.iec.ch/61850/2003/SCL" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://www.iec.ch/61850/2003/SCL c:\SCL\SCL.xsd">
<Header id="9-2LE-Spec" nameStructure="IEDName" version="0.7" revision="1"/>
<Substation name="RTDS">
<VoltageLevel name="230kV">
<Bay name="5L1">
<ConductingEquipment name="Inn" type="CTR">
<SubEquipment name="A" phase="A">
<LNode iedName="GTNETSV1" ldInst="MU00" lnClass="TCTR" lnInst="1"/>
</SubEquipment>
<SubEquipment name="B" phase="B">
<LNode iedName="GTNETSV1" ldInst="MU00" lnClass="TCTR" lnInst="2"/>
</SubEquipment>
<SubEquipment name="C" phase="C">
<LNode iedName="GTNETSV1" ldInst="MU00" lnClass="TCTR" lnInst="3"/>
</SubEquipment>
<SubEquipment name="N" phase="N">
<LNode iedName="GTNETSV1" ldInst="MU00" lnClass="TCTR" lnInst="4"/>
</SubEquipment>
</ConductingEquipment>
<ConductingEquipment name="Unn" type="VTR">
<SubEquipment name="A" phase="A">
<LNode iedName="GTNETSV1" ldInst="MU00" lnClass="TVTR" lnInst="1"/>
</SubEquipment>
<SubEquipment name="B" phase="B">
<LNode iedName="GTNETSV1" ldInst="MU00" lnClass="TVTR" lnInst="2"/>
</SubEquipment>
<SubEquipment name="C" phase="C">
<LNode iedName="GTNETSV1" ldInst="MU00" lnClass="TVTR" lnInst="3"/>
</SubEquipment>
<SubEquipment name="N" phase="N">
<LNode iedName="GTNETSV1" ldInst="MU00" lnClass="TVTR" lnInst="4"/>
</SubEquipment>
</ConductingEquipment>
137
</Bay>
</VoltageLevel>
</Substation>
<Communication>
<SubNetwork name="NONE" type="8-MMS">
<ConnectedAP iedName="GTNETSV1" apName="P1">
<SMV ldInst="MU00" cbName="MSVCB01">
<Address>
<P type="VLAN-ID">000</P>
<P type="VLAN-PRIORITY">4</P>
<P type="MAC-Address">01-0C-CD-04-01-1D</P>
<P type="APPID">4000</P>
</Address>
</SMV>
</ConnectedAP>
</SubNetwork>
</Communication>
<IED name="GTNETSV1" manufacturer="RTDS Technologies Inc." >
<Services>
<ConfDataSet maxAttributes="100" max="2" modify="0"/>
<ReadWrite/>
</Services>
<AccessPoint name="P1">
<Server>
<Authentication/>
<LDevice inst="MU00">
<LN0 lnType="9-2LELLN0" lnClass="LLN0" inst="">
<DataSet name="PhsMeas1">
<FCDA ldInst="MU00" lnClass="TCTR" lnInst="1" doName="Amp" daName="instMag.i" fc="MX"/>
<FCDA ldInst="MU00" lnClass="TCTR" lnInst="1" doName="Amp" daName="q" fc="MX"/>
<FCDA ldInst="MU00" lnClass="TCTR" lnInst="2" doName="Amp" daName="instMag.i" fc="MX"/>
<FCDA ldInst="MU00" lnClass="TCTR" lnInst="2" doName="Amp" daName="q" fc="MX"/>
<FCDA ldInst="MU00" lnClass="TCTR" lnInst="3" doName="Amp" daName="instMag.i" fc="MX"/>
<FCDA ldInst="MU00" lnClass="TCTR" lnInst="3" doName="Amp" daName="q" fc="MX"/>
<FCDA ldInst="MU00" lnClass="TCTR" lnInst="4" doName="Amp" daName="instMag.i" fc="MX"/>
<FCDA ldInst="MU00" lnClass="TCTR" lnInst="4" doName="Amp" daName="q" fc="MX"/>
<FCDA ldInst="MU00" lnClass="TVTR" lnInst="1" doName="Vol" daName="instMag.i" fc="MX"/>
<FCDA ldInst="MU00" lnClass="TVTR" lnInst="1" doName="Vol" daName="q" fc="MX"/>
<FCDA ldInst="MU00" lnClass="TVTR" lnInst="2" doName="Vol" daName="instMag.i" fc="MX"/>
<FCDA ldInst="MU00" lnClass="TVTR" lnInst="2" doName="Vol" daName="q" fc="MX"/>
<FCDA ldInst="MU00" lnClass="TVTR" lnInst="3" doName="Vol" daName="instMag.i" fc="MX"/>
<FCDA ldInst="MU00" lnClass="TVTR" lnInst="3" doName="Vol" daName="q" fc="MX"/>
<FCDA ldInst="MU00" lnClass="TVTR" lnInst="4" doName="Vol" daName="instMag.i" fc="MX"/>
<FCDA ldInst="MU00" lnClass="TVTR" lnInst="4" doName="Vol" daName="q" fc="MX"/>
</DataSet> <SampledValueControl name="MSVCB01" datSet="PhsMeas1" smvID="EXER1MU0001" smpRate="80" nofASDU="1" confRev="0">
<SmvOpts refreshTime="false" sampleSynchronized="true" sampleRate="false" security="false" dataRef="false"/>
</SampledValueControl>
</LN0>
<LN desc="" lnType="9-2LELPHD" lnClass="LPHD" inst="1"/>
<LN lnType="9-2LETCTR" lnClass="TCTR" inst="1"/>
<LN lnType="9-2LETCTR" lnClass="TCTR" inst="2"/>
<LN lnType="9-2LETCTR" lnClass="TCTR" inst="3"/>
<LN lnType="9-2LETCTR" lnClass="TCTR" inst="4"/>
<LN lnType="9-2LETVTR" lnClass="TVTR" inst="1"/>
<LN lnType="9-2LETVTR" lnClass="TVTR" inst="2"/>
<LN lnType="9-2LETVTR" lnClass="TVTR" inst="3"/>
<LN lnType="9-2LETVTR" lnClass="TVTR" inst="4"/>
</LDevice>
</Server>
</AccessPoint>
</IED>
<DataTypeTemplates>
<LNodeType id="9-2LELPHD" lnClass="LPHD">
<DO name="PhyNam" type="9-2LEPHYNAM"/>
<DO name="PhyHealth" type="9-2LEINS_HEALTH"/>
<DO name="Proxy" type="9-2LEPROXY"/>
</LNodeType>
<LNodeType id="9-2LELLN0" lnClass="LLN0">
<DO name="Mod" type="9-2LEINC"/>
<DO name="Beh" type="9-2LEINS_BEH"/>
<DO name="Health" type="9-2LEINS_HEALTH"/>
<DO name="NamPlt" type="9-2LELN0LPL"/>
</LNodeType>
<LNodeType id="9-2LETCTR" lnClass="TCTR">
<DO name="Mod" type="9-2LEINC"/>
<DO name="Beh" type="9-2LEINS_BEH"/>
<DO name="Health" type="9-2LEINS_HEALTH"/>
138
<DO name="NamPlt" type="9-2LELNLPL"/>
<DO name="Amp" type="9-2LESAVAmp"/>
</LNodeType>
<LNodeType id="9-2LETVTR" lnClass="TVTR">
<DO name="Mod" type="9-2LEINC"/>
<DO name="Beh" type="9-2LEINS_BEH"/>
<DO name="Health" type="9-2LEINS_HEALTH"/>
<DO name="NamPlt" type="9-2LELNLPL"/>
<DO name="Vol" type="9-2LESAVVol"/>
</LNodeType>
<DOType id="9-2LEINS_BEH" cdc="INS">
<DA name="stVal" bType="Enum" type="Beh" fc="ST" dchg="true">
<Val>on</Val>
</DA>
<DA name="q" bType="Quality" fc="ST" valKind="RO" qchg="true"/>
<DA name="t" bType="Timestamp" fc="ST" valKind="RO"/>
</DOType>
<DOType id="9-2LEINS_HEALTH" cdc="INS">
<DA name="stVal" bType="Enum" type="HealthKind" fc="ST" dchg="true">
<Val>Ok</Val>
</DA>
<DA name="q" bType="Quality" fc="ST" valKind="RO" qchg="true"/>
<DA name="t" bType="Timestamp" fc="ST" valKind="RO"/>
</DOType>
<DOType id="9-2LESAVAmp" cdc="SAV">
<DA name="instMag" bType="Struct" type="9-2LEAV" fc="MX"/>
<DA name="q" bType="Quality" fc="MX" valKind="RO" qchg="true"/>
<DA name="sVC" bType="Struct" type="9-2LEsVCAmp" fc="CF"/>
</DOType>
<DOType id="9-2LESAVVol" cdc="SAV">
<DA name="instMag" bType="Struct" type="9-2LEAV" fc="MX"/>
<DA name="q" bType="Quality" fc="MX" valKind="RO" qchg="true"/>
<DA name="sVC" bType="Struct" type="9-2LEsVCVol" fc="CF"/>
</DOType>
<DOType id="9-2LEINC" cdc="INC">
<DA name="stVal" fc="ST" bType="Enum" type="Mod" dchg="true">
<Val>on</Val>
</DA>
<DA name="q" bType="Quality" fc="ST" valKind="RO" qchg="true"/>
<DA name="t" bType="Timestamp" fc="ST" valKind="RO"/>
<DA name="ctlModel" fc="CF" bType="Enum" type="CtlModelKind">
<Val>status-only</Val>
</DA>
</DOType>
<DOType id="9-2LELN0LPL" cdc="LPL">
<DA name="vendor" bType="VisString255" fc="DC"/>
<DA name="swRev" bType="VisString255" fc="DC"/>
<DA name="d" bType="VisString255" fc="DC"/>
<DA name="configRev" bType="VisString255" fc="DC"/>
<DA name="ldNs" bType="VisString255" fc="EX">
<Val>IEC 61850-7-4:2003</Val>
</DA>
</DOType>
<DOType id="9-2LELNLPL" cdc="LPL">
<DA name="vendor" bType="VisString255" fc="DC"/>
<DA name="swRev" bType="VisString255" fc="DC"/>
<DA name="d" bType="VisString255" fc="DC"/>
</DOType>
<DOType id="9-2LEPROXY" cdc="SPS">
<DA name="stVal" bType="BOOLEAN" dchg="true" fc="ST">
<Val>false</Val>
</DA>
<DA name="q" bType="Quality" fc="ST" valKind="RO" qchg="true"/>
<DA name="t" bType="Timestamp" fc="ST" valKind="RO"/>
</DOType>
<DOType id="9-2LEPHYNAM" cdc="DPL">
<DA name="vendor" bType="VisString255" fc="DC"/>
<DA name="hwRev" bType="VisString255" fc="DC"/>
<DA name="swRev" bType="VisString255" fc="DC"/>
<DA name="serNum" bType="VisString255" fc="DC"/>
<DA name="model" bType="VisString255" fc="DC"/>
<DA name="location" bType="VisString255" fc="DC"/>
<DA name="cdcNs" bType="VisString255" fc="EX"/>
</DOType>
<DAType id="9-2LEAV">
<BDA name="i" bType="INT32"/>
</DAType>
<DAType id="9-2LEsVCAmp">
139
<BDA name="scaleFactor" bType="FLOAT32">
<Val>0.001</Val>
</BDA>
<BDA name="offset" bType="FLOAT32">
<Val>0</Val>
</BDA>
</DAType>
<DAType id="9-2LEsVCVol">
<BDA name="scaleFactor" bType="FLOAT32">
<Val>0.01</Val>
</BDA>
<BDA name="offset" bType="FLOAT32">
<Val>0</Val>
</BDA>
</DAType>
<EnumType id="Beh">
<EnumVal ord="1">on</EnumVal>
<EnumVal ord="2">blocked</EnumVal>
<EnumVal ord="3">test</EnumVal>
<EnumVal ord="4">test/blocked</EnumVal>
<EnumVal ord="5">off</EnumVal>
</EnumType>
<EnumType id="Mod">
<EnumVal ord="1">on</EnumVal>
<EnumVal ord="2">blocked</EnumVal>
<EnumVal ord="3">test</EnumVal>
<EnumVal ord="4">test/blocked</EnumVal>
<EnumVal ord="5">off</EnumVal>
</EnumType>
<EnumType id="HealthKind">
<EnumVal ord="1">Ok</EnumVal>
<EnumVal ord="2">Warning</EnumVal>
<EnumVal ord="3">Alarm</EnumVal>
</EnumType>
<EnumType id="CtlModelKind">
<EnumVal ord="0">status-only</EnumVal>
<EnumVal ord="1">direct-with-normal-security</EnumVal>
<EnumVal ord="2">sbo-with-normal-security</EnumVal>
<EnumVal ord="3">direct-with-enhanced-security</EnumVal>
<EnumVal ord="4">sbo-with-enhanced-security</EnumVal>
</EnumType>
</DataTypeTemplates>
</SCL>
III. CID file of the Line Protection IED
<?xml version="1.0" encoding="UTF-8"?>
<!--This file is generated using KALKI SCL Manager IEC61850 Configuration Tool (www.kalkitech.com)-->
<SCL xmlns="http://www.iec.ch/61850/2003/SCL" xmlns:xsd="http://www.w3.org/2001/XMLSchema">
<Header id="" version="4.0.0 Beta" revision="" toolID="" nameStructure="IEDName" />
<Communication>
<SubNetwork name="subnetwork1" type="8-MMS">
<Text>Station bus</Text>
<BitRate unit="b/s">10</BitRate>
<ConnectedAP iedName="smv" apName="accessPoint1">
<Address>
<P type="IP">192.168.100.81</P>
<P type="IP-SUBNET">255.255.255.0</P>
<P type="IP-GATEWAY">192.168.100.1</P>
<P type="OSI-TSEL">0001</P>
<P type="OSI-PSEL">00000001</P>
<P type="OSI-SSEL">0001</P>
</Address>
<GSE ldInst="lDevice1" cbName="GoosePublisher">
<Address>
<P type="MAC-Address">01-0C-CD-01-01-CD</P>
<P type="APPID">01</P>
<P type="VLAN-ID">0</P>
<P type="VLAN-PRIORITY">4</P>
140
</Address>
</GSE>
</ConnectedAP>
</SubNetwork>
</Communication>
<IED name="smv">
<Services>
<DynAssociation />
<GetDirectory />
<GetDataObjectDefinition />
<GetDataSetValue />
<DataSetDirectory />
<ReadWrite />
<GetCBValues />
<ConfLNs fixPrefix="true" fixLnInst="true" />
<GOOSE max="5" />
<GSSE max="5" />
<FileHandling />
<GSEDir />
<TimerActivatedControl />
</Services>
<AccessPoint name="accessPoint1">
<Server>
<Authentication />
<LDevice inst="lDevice1">
<Private type="SMVMapping">GTNETSV1MU02/LLN0$MS$MSVCB01|PhsMeas1|EXER1MU0001|01-0C-CD-04-01-
1D|FF-FF-FF-FF-FF-FF|
|NumDatasetVars|16|
|Primitive|smvlDevice1/TCTR1$MX$Amp$instMag$i|GTNETSV1\MU02.MSVCB01\SV_Publisher_Data\MU02/TCTR1$MX$Amp$instMag$i|
|Primitive|smvlDevice1/TCTR1$MX$Amp$q|GTNETSV1\MU02.MSVCB01\SV_Publisher_Data\MU02/TCTR1$MX$Amp$q|
|Primitive|smvlDevice1/TCTR2$MX$Amp$instMag$i|GTNETSV1\MU02.MSVCB01\SV_Publisher_Data\MU02/TCTR2$MX$Amp$instMag$i|
|Primitive|smvlDevice1/TCTR2$MX$Amp$q|GTNETSV1\MU02.MSVCB01\SV_Publisher_Data\MU02/TCTR2$MX$Amp$q|
|Primitive|smvlDevice1/TCTR3$MX$Amp$instMag$i|GTNETSV1\MU02.MSVCB01\SV_Publisher_Data\MU02/TCTR3$MX$Amp$instMag$i|
|Primitive|smvlDevice1/TCTR3$MX$Amp$q|GTNETSV1\MU02.MSVCB01\SV_Publisher_Data\MU02/TCTR3$MX$Amp$q|
|Primitive|smvlDevice1/TCTR4$MX$Amp$instMag$i|GTNETSV1\MU02.MSVCB01\SV_Publisher_Data\MU02/TCTR4$MX$Amp$instMag$i|
|Primitive|smvlDevice1/TCTR4$MX$Amp$q|GTNETSV1\MU02.MSVCB01\SV_Publisher_Data\MU02/TCTR4$MX$Amp$q|
|Primitive|smvlDevice1/TVTR1$MX$Vol$instMag$i|GTNETSV1\MU02.MSVCB01\SV_Publisher_Data\MU02/TCTR1$MX$Vol$instMag$i|
|Primitive|smvlDevice1/TVTR1$MX$Vol$q|GTNETSV1\MU02.MSVCB01\SV_Publisher_Data\MU02/TCTR1$MX$Vol$q|
|Primitive|smvlDevice1/TVTR2$MX$Vol$instMag$i|GTNETSV1\MU02.MSVCB01\SV_Publisher_Data\MU02/TCTR2$MX$Vol$instMag$i|
|Primitive|smvlDevice1/TVTR2$MX$Vol$q|GTNETSV1\MU02.MSVCB01\SV_Publisher_Data\MU02/TCTR2$MX$Vol$q|
|Primitive|smvlDevice1/TVTR3$MX$Vol$instMag$i|GTNETSV1\MU02.MSVCB01\SV_Publisher_Data\MU02/TCTR3$MX$Vol$instMag$i|
|Primitive|smvlDevice1/TVTR3$MX$Vol$q|GTNETSV1\MU02.MSVCB01\SV_Publisher_Data\MU02/TCTR3$MX$Vol$q|
|Primitive|smvlDevice1/TVTR4$MX$Vol$instMag$i|GTNETSV1\MU02.MSVCB01\SV_Publisher_Data\MU02/TCTR4$MX$Vol$instMag$i|
|Primitive|smvlDevice1/TVTR4$MX$Vol$q|GTNETSV1\MU02.MSVCB01\SV_Publisher_Data\MU02/TCTR4$MX$Vol$q|
</Private>
<LN0 lnClass="LLN0" lnType="LLN04" inst="">
<DataSet name="Dataset1" desc="Goose publisher dataset">
<FCDA ldInst="lDevice1" lnClass="GGIO" fc="ST" lnInst="1" prefix="GSPUB_" doName="Ind1" daName="stVal" />
<FCDA ldInst="lDevice1" lnClass="GGIO" fc="ST" lnInst="1" prefix="GSPUB_" doName="Ind2" daName="stVal" />
<FCDA ldInst="lDevice1" lnClass="GGIO" fc="ST" lnInst="1" prefix="GSPUB_" doName="Ind3" daName="stVal" />
<FCDA ldInst="lDevice1" lnClass="GGIO" fc="ST" lnInst="1" prefix="GSPUB_" doName="Ind4" daName="stVal" />
<FCDA ldInst="lDevice1" lnClass="GGIO" fc="ST" lnInst="1" prefix="GSPUB_" doName="Ind5" daName="stVal" />
</DataSet>
<DOI name="Mod">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Beh">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Health">
<DAI name="stVal">
<Val>Ok</Val>
141
</DAI>
</DOI>
<GSEControl name="GoosePublisher" appID="UoMg1" type="GOOSE" datSet="Dataset1" confRev="1" />
</LN0>
<LN lnClass="LPHD" lnType="LPHD4" inst="1" prefix="">
<DOI name="PhyHealth">
<DAI name="stVal">
<Val>Ok</Val>
</DAI>
</DOI>
</LN>
<LN lnClass="GGIO" lnType="GGIO1" inst="1" prefix="GSPUB_">
<DOI name="Mod">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Beh">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Health">
<DAI name="stVal">
<Val>Ok</Val>
</DAI>
</DOI>
</LN>
<LN lnClass="TCTR" lnType="TCTR1" inst="1" prefix="">
<DOI name="Mod">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Beh">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Health">
<DAI name="stVal">
<Val>Ok</Val>
</DAI>
</DOI>
</LN>
<LN lnClass="TVTR" lnType="TVTR1" inst="1" prefix="">
<DOI name="Mod">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Beh">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Health">
<DAI name="stVal">
<Val>Ok</Val>
</DAI>
</DOI>
</LN>
<LN lnClass="TCTR" lnType="TCTR2" inst="2" prefix="">
<DOI name="Mod">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Beh">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Health">
<DAI name="stVal">
<Val>Ok</Val>
</DAI>
</DOI>
142
</LN>
<LN lnClass="TCTR" lnType="TCTR3" inst="3" prefix="">
<DOI name="Mod">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Beh">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Health">
<DAI name="stVal">
<Val>Ok</Val>
</DAI>
</DOI>
</LN>
<LN lnClass="TCTR" lnType="TCTR3" inst="4" prefix="">
<DOI name="Mod">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Beh">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Health">
<DAI name="stVal">
<Val>Ok</Val>
</DAI>
</DOI>
</LN>
<LN lnClass="TCTR" lnType="TCTR3" inst="5" prefix="">
<DOI name="Mod">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Beh">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Health">
<DAI name="stVal">
<Val>Ok</Val>
</DAI>
</DOI>
</LN>
<LN lnClass="TCTR" lnType="TCTR3" inst="6" prefix="">
<DOI name="Mod">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Beh">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Health">
<DAI name="stVal">
<Val>Ok</Val>
</DAI>
</DOI>
</LN>
<LN lnClass="TCTR" lnType="TCTR3" inst="7" prefix="">
<DOI name="Mod">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Beh">
<DAI name="stVal">
<Val>on</Val>
143
</DAI>
</DOI>
<DOI name="Health">
<DAI name="stVal">
<Val>Ok</Val>
</DAI>
</DOI>
</LN>
<LN lnClass="TCTR" lnType="TCTR3" inst="8" prefix="">
<DOI name="Mod">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Beh">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Health">
<DAI name="stVal">
<Val>Ok</Val>
</DAI>
</DOI>
</LN>
<LN lnClass="TVTR" lnType="TVTR2" inst="2" prefix="">
<DOI name="Mod">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Beh">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Health">
<DAI name="stVal">
<Val>Ok</Val>
</DAI>
</DOI>
</LN>
<LN lnClass="TVTR" lnType="TVTR2" inst="3" prefix="">
<DOI name="Mod">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Beh">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Health">
<DAI name="stVal">
<Val>Ok</Val>
</DAI>
</DOI>
</LN>
<LN lnClass="TVTR" lnType="TVTR2" inst="4" prefix="">
<DOI name="Mod">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Beh">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Health">
<DAI name="stVal">
<Val>Ok</Val>
</DAI>
</DOI>
</LN>
<LN lnClass="TVTR" lnType="TVTR2" inst="5" prefix="">
<DOI name="Mod">
144
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Beh">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Health">
<DAI name="stVal">
<Val>Ok</Val>
</DAI>
</DOI>
</LN>
<LN lnClass="TVTR" lnType="TVTR2" inst="6" prefix="">
<DOI name="Mod">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Beh">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Health">
<DAI name="stVal">
<Val>Ok</Val>
</DAI>
</DOI>
</LN>
<LN lnClass="TVTR" lnType="TVTR2" inst="7" prefix="">
<DOI name="Mod">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Beh">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Health">
<DAI name="stVal">
<Val>Ok</Val>
</DAI>
</DOI>
</LN>
<LN lnClass="TVTR" lnType="TVTR2" inst="8" prefix="">
<DOI name="Mod">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Beh">
<DAI name="stVal">
<Val>on</Val>
</DAI>
</DOI>
<DOI name="Health">
<DAI name="stVal">
<Val>Ok</Val>
</DAI>
</DOI>
</LN>
</LDevice>
</Server>
</AccessPoint>
</IED>
<DataTypeTemplates>
<LNodeType id="LLN04" lnClass="LLN0">
<DO name="Mod" type="INC_0" />
<DO name="Beh" type="INS_0" />
<DO name="Health" type="INS_1" />
<DO name="NamPlt" type="LPL_2_NamPlt" />
</LNodeType>
<LNodeType id="LPHD4" lnClass="LPHD">
145
<DO name="PhyNam" type="DPL_1_PhyNam" />
<DO name="PhyHealth" type="INS_2_PhyHealth" />
<DO name="Proxy" type="SPS_1" />
</LNodeType>
<LNodeType id="GGIO1" lnClass="GGIO">
<DO name="Mod" type="INC_0" />
<DO name="Beh" type="INS_0" />
<DO name="Health" type="INS_1" />
<DO name="Ind1" type="SPS_1" />
<DO name="Ind2" type="SPS_1" />
<DO name="Ind3" type="SPS_1" />
<DO name="Ind4" type="SPS_1" />
<DO name="Ind5" type="SPS_1" />
</LNodeType>
<LNodeType id="TCTR1" lnClass="TCTR">
<DO name="Mod" type="INC_0" />
<DO name="Beh" type="INS_0" />
<DO name="Health" type="INS_1" />
<DO name="NamPlt" type="LPL_1" />
<DO name="Amp" type="SAV_1_Amp" />
</LNodeType>
<LNodeType id="TVTR1" lnClass="TVTR">
<DO name="Mod" type="INC_0" />
<DO name="Beh" type="INS_0" />
<DO name="Health" type="INS_1" />
<DO name="NamPlt" type="LPL_1" />
<DO name="Vol" type="SAV_1_Amp" />
</LNodeType>
<LNodeType id="TCTR2" lnClass="TCTR">
<DO name="Mod" type="INC_0" />
<DO name="Beh" type="INS_0" />
<DO name="Health" type="INS_1" />
<DO name="NamPlt" type="LPL_1" />
<DO name="Amp" type="SAV_1_Amp" />
</LNodeType>
<LNodeType id="TCTR3" lnClass="TCTR">
<DO name="Mod" type="INC_0" />
<DO name="Beh" type="INS_0" />
<DO name="Health" type="INS_1" />
<DO name="NamPlt" type="LPL_1" />
<DO name="Amp" type="SAV_1_Amp" />
</LNodeType>
<LNodeType id="TVTR2" lnClass="TVTR">
<DO name="Mod" type="INC_0" />
<DO name="Beh" type="INS_0" />
<DO name="Health" type="INS_1" />
<DO name="NamPlt" type="LPL_1" />
<DO name="Vol" type="SAV_1_Amp" />
</LNodeType>
<DOType id="INC_0" cdc="INC">
<DA name="stVal" type="Mod" bType="Enum" fc="ST" dchg="true" />
<DA name="q" bType="Quality" fc="ST" qchg="true" />
<DA name="t" bType="Timestamp" fc="ST" />
<DA name="ctlModel" type="ctlModel" bType="Enum" fc="CF" />
</DOType>
<DOType id="INS_0" cdc="INS">
<DA name="stVal" type="Beh" bType="Enum" fc="ST" />
<DA name="q" bType="Quality" fc="ST" />
<DA name="t" bType="Timestamp" fc="ST" />
</DOType>
<DOType id="INS_1" cdc="INS">
<DA name="stVal" type="Health" bType="Enum" fc="ST" />
<DA name="q" bType="Quality" fc="ST" />
<DA name="t" bType="Timestamp" fc="ST" />
</DOType>
<DOType id="LPL_2_NamPlt" cdc="LPL">
<DA name="vendor" bType="VisString255" fc="DC" />
<DA name="swRev" bType="VisString255" fc="DC" />
<DA name="d" bType="VisString255" fc="DC" />
<DA name="configRev" bType="VisString255" fc="DC" />
<DA name="ldNs" bType="VisString255" fc="EX" />
</DOType>
<DOType id="DPL_1_PhyNam" cdc="DPL">
<DA name="vendor" bType="VisString255" fc="DC" />
</DOType>
<DOType id="INS_2_PhyHealth" cdc="INS">
<DA name="stVal" type="PhyHealth" bType="Enum" fc="ST" dchg="true" />
<DA name="q" bType="Quality" fc="ST" qchg="true" />
146
<DA name="t" bType="Timestamp" fc="ST" />
</DOType>
<DOType id="SPS_1" cdc="SPS">
<DA name="stVal" bType="BOOLEAN" fc="ST" dchg="true" />
<DA name="q" bType="Quality" fc="ST" qchg="true" />
<DA name="t" bType="Timestamp" fc="ST" />
</DOType>
<DOType id="LPL_1" cdc="LPL">
<DA name="vendor" bType="VisString255" fc="DC" />
<DA name="swRev" bType="VisString255" fc="DC" />
<DA name="d" bType="VisString255" fc="DC" />
</DOType>
<DOType id="SAV_1_Amp" cdc="SAV">
<DA name="instMag" type="AnalogueValue_1" bType="Struct" fc="MX" />
<DA name="q" bType="Quality" fc="MX" qchg="true" />
</DOType>
<DAType id="AnalogueValue_1">
<BDA name="i" bType="INT32" />
</DAType>
<EnumType id="Mod">
<EnumVal ord="1">on</EnumVal>
<EnumVal ord="2">blocked</EnumVal>
<EnumVal ord="3">test</EnumVal>
<EnumVal ord="4">test/blocked</EnumVal>
<EnumVal ord="5">off</EnumVal>
</EnumType>
<EnumType id="ctlModel">
<EnumVal ord="0">status-only</EnumVal>
<EnumVal ord="1">direct-with-normal-security</EnumVal>
<EnumVal ord="2">sbo-with-normal-security</EnumVal>
<EnumVal ord="3">direct-with-enhanced-security</EnumVal>
<EnumVal ord="4">sbo-with-enhanced-security</EnumVal>
</EnumType>
<EnumType id="Beh">
<EnumVal ord="1">on</EnumVal>
<EnumVal ord="2">blocked</EnumVal>
<EnumVal ord="3">test</EnumVal>
<EnumVal ord="4">test-blocked</EnumVal>
<EnumVal ord="5">off</EnumVal>
</EnumType>
<EnumType id="Health">
<EnumVal ord="1">Ok</EnumVal>
<EnumVal ord="2">Warning</EnumVal>
<EnumVal ord="3">Alarm</EnumVal>
</EnumType>
<EnumType id="PhyHealth">
<EnumVal ord="1">Ok</EnumVal>
<EnumVal ord="2">Warning</EnumVal>
<EnumVal ord="3">Alarm</EnumVal>
</EnumType>
</DataTypeTemplates>
</SCL>
Appendix B
I. Transmission Line Parameters
Following are the main parameters used in the transmission line model
Model: Frequency Dependent phase model
Line length: 200 km
Ground Resistivity: 100 Ωm
Low frequency: 0.1 Hz
High frequency: 1 MHz
Steady state frequency: 60 Hz
Number of conductors on tower: 3
Number of ground wires on tower: 2
11.2 m
22.5 m
14.4 m 14.4 m
43.7 m
Conductors
Ground wires
Figure (a): Tower
configuration
148
Sub-conductor radius: 1.666 cm
DC resistance per sub-conductor: 0.0472 Ω/km
Shunt conductance: 1 × 10−11 mho/m (S/m)
Sub-conductors per bundle: 2
Sub-conductor spacing: 40 cm
Ground wire radius: 0.489 cm
DC resistance per wire: 1.463 Ω/km
Sag at mid-span: 15 m (for both conductor and ground wire)
II. Series Capacitor Parameters
Figure (b): Thyristor protected
series capacitor
149
Series Capacitance: 80 µF
Arrestor parameters
Discharge current: 10 kA
Discharge voltage: 338 kV
Energy decay time constant: 0.5 s
Parameters of the parallel thyristor branch
Rth: 1000 MΩ
Lth: 1000 MH
Rt2: 1 Ω
Rs: 1000 MΩ
Cs: 1 pF
Value on resistance: 10 µΩ
Value off resistance: 1 MΩ
Parameters of the parallel thyristor branch
R1: 1 Ω
R2: 0.01 Ω
L1: 2 mH
These are default values from the RTDS library component.
150
Appendix C
Filter coefficients for the mother wavelet “Daubechies 4
(db4)”.
Coefficients for the decomposition high-pass filter h[n].
h[0] = -0.23037781330885523
h[1] = 0.7148465705525415
h[2] = -0.6308807679295904
h[3] = -0.02798376941698385
h[4] = 0.18703481171888114
h[5] = 0.030841381835986965
h[6] = -0.032883011666982945
h[7] = -0.010597401784997278
0 1 2 3 4 5 6 7 8-0.4
-0.2
0
0.2
0.4
0.6
0.8
1
1.2db4 : phi
0 1 2 3 4 5 6 7 8-1
-0.5
0
0.5
1
1.5
(a)
Figure (c): “db4” mother
wavelet function
151
Appendix D
I. Behaviour of traveling waves at a junction on a
transmission line
Figure (d): Travelling wave at a junction
Above Figure (d) illustrates the behaviour of a travelling wave at a junction
formed by the connection of two transmission lines. The original wave, propagating
through transmission line 1 towards the junction J, is called the incident wave. A
portion of this wave is reflected back at the junction (the reflected wave) and the rest
of it is transmitted into the second transmission line (the transmitted wave). Voltage
and current magnitudes of the incident, the reflected and the transmitted waves are
denotes by subscripts I, R and T, respectively. These waves comply with the
differential equation of the line as well as Kirchhoff's current and voltage laws.
J
Incident wave
Reflected wave
Transmitted wave
ET, IT
ER, IR
EI, II
Z2
Z1
Line 1
Line 2
152
Moreover, the surge impedances of the two transmission lines, Z1 and Z2, relates the
voltages of the lines to the respective currents. With this comprehension, following
equations can be written,
EI + ER = ET and II + IR = IT
EI = Z1II, ET = Z2IT and ER = −Z1IR
The transmitted and reflected voltages can be obtained by rearranging the above
equations as follows:
ET =2Z2
Z1 + Z2. EI and ER =
Z2 − Z1
Z1 + Z2. EI
Similarly, transmitted and reflected currents are given by:
IT =2Z1
Z1 + Z2. II and IR =
Z1 − Z2
Z1 + Z2. II
Hence, the transmission and reflection coefficients, α and β, respectively, for voltages
and currents can be expressed as,
αE =2Z2
Z1 + Z2 and βE =
Z2 − Z1
Z1 + Z2
αI =2Z1
Z1 + Z2 and βI =
Z1 − Z2
Z1 + Z2
153
Case I
Following Figure (e) depicts a transmission line linking two AC networks; a
system similar to the one used for the transient protection scheme of this work. Now
let’s consider the behaviour of a current surge from system 1, at junction 1 (bus 1),
occurring due to a solid fault in the transmission line. A fault occurring in the
transmission line would effectively divide the line into two sections with respective
surge impedances Z2 and Z3 as shown in the figure below. The equivalent surge
impedance of system 1 is Z1, whereas that of the line section from bus 1 to the location
of the fault is Z2. The first transient observed by the CT at point A due to this fault
would be the transmitted current wave at junction 1.
Figure (e): Fault on the line; an internal fault
The transmitted wave in terms of the incident wave is,
IT =2Z1
Z1 + Z2. II =
2(Z1 Z2⁄ )
(Z1 Z2⁄ ) + 1. II
From this it can be said that for all combinations of Z1 and Z2, point A shall
observe a fraction of the incident current wave, with its polarity unchanged (phase
angles of Z1 and Z2 are typically quite close to each other, therefore, it is highly
System
Equivalent 1 System
Equivalent 2
Bus 2 Bus 1
Fault
B A
Z1 Z
2 Z
3
154
unlikely that 2Z1
Z1+Z2 would enforce a phase angle shift on II to reverse its polarity).
However, the magnitude of the transmitted wave depends upon the ratio Z1/Z2. The
same argument can be extended to the behaviour of the current waves at junction 2
(bus 2) as well. Therefore, in theory, the two current measurement points shall
observe initial transients of the same polarity for faults occurring on the line for all
systems. The exact polarity depends on direction of change of the initial fault current,
with respect to the pre-fault steady state current, which is determined by the fault
inception angle.
Case II
Now let’s consider the behaviour of a travelling wave (current) for a solid fault
at Bus 1. Now, the equivalent surge impedance of systems 1 and 2 are Z1 and Z3,
respectively, and that of the transmission line is Z2.
Figure (f): Fault on the bus; an external fault
Now the junction at bus 1 is on a short-circuit (therefore, the transmission and
reflection coefficients are 2 and 1, respectively). However, since the line is short-
circuited at bus 1, the current surge coming from system 1 will not go beyond bus 1
and hence, the current measurement at point A will not observe it.
System
Equivalent 1 System
Equivalent 2
Bus 2 Bus 1
Fault
B A
Z1 Z
2 Z
3
155
On the other hand, current surge coming from system 2, at bus 2 follows a
similar behaviour as explained in Case I. Thus, the first transient observed by both
the points A and B would be the transmitted wave at bus 2 of the current surge
coming from system 2. The observed the polarity of this wave at the two points would
obviously be opposite to each other due to opposite directions of current measurement.
This transmitted wave will also be short-circuited at bus 1.
II. Analysis of Wave-front polarities for a R-L circuit
𝑉1∠𝛿
Bus 2 Bus 1
I1
Fault occurring at
t = t0
B A 𝑉2∠0
I2
Z
Figure (g): Pre-Fault Circuit
Z2 Z1
𝑉1∠𝛿
Bus 2 Bus 1
If1
B A 𝑉2∠0
If2
Figure (h): Post-Fault circuits
156
Let’s consider a system, where two ideal AC voltage sources of frequency f, with
a phase angle difference of 𝛿, are connected to a R-L impedance. The magnitude of
this impedance is Z. A solid fault occurring at t = t0 divides the system into two
circuits, each with a source and an impedance, as shown in the above figure. The
magnitude of these two impedance sections are Z1 and Z2, respectively. Let’s take,
𝑍1 = √𝑅12 + (𝜔. 𝐿1)2 , 𝑎𝑛𝑑 𝑍2 = √𝑅2
2 + (𝜔. 𝐿2)2
; Where 𝜔 = 2𝜋𝑓. Then the impedance angle 𝜙 is given by,
𝜙 = 𝑡𝑎𝑛−1 (𝜔. 𝐿1
𝑅1) = 𝑡𝑎𝑛−1 (
𝜔. 𝐿2
𝑅2)
The instantaneous, steady state current prior to the fault, can be written as (current
going from bus 1 to bus2 is taken as positive),
𝐼1 = −𝐼2 =𝑉1
𝑍1 + 𝑍2sin(𝜔𝑡 + 𝛿 − 𝜙) −
𝑉2
𝑍1 + 𝑍2sin(𝜔𝑡 − 𝜙)
From the transient analysis of the two circuits, it can be proven that the fault
currents of the two circuits are given by,
𝐼𝑓1 = 𝐼0 +𝑉1
𝑍1sin(𝜙 − 𝛼 − 𝛿) 𝑒
−𝑅1𝐿1
(𝑡−𝑡0)+
𝑉1
𝑍1sin(𝜔𝑡 + 𝛿 − 𝜙)
; Where I0 is the steady state current at t = t0 and 𝛼 = 𝜔𝑡0 (𝛼 is also the inception
angle of the fault)
𝐼𝑓1 = 𝑉1
𝑍1 + 𝑍2sin(𝛼 + 𝛿 − 𝜙) −
𝑉2
𝑍1 + 𝑍2sin(𝛼 − 𝜙) +
𝑉1
𝑍1sin(𝜙 − 𝛼 − 𝛿) 𝑒
−𝑅1𝐿1
(𝑡−𝑡0)
+𝑉1
𝑍1sin(𝜔𝑡 + 𝛿 − 𝜙)
157
Similarly,
𝐼𝑓2 = −𝐼0 +𝑉2
𝑍2sin(𝜙 − 𝛼) 𝑒
−𝑅2𝐿2
(𝑡−𝑡0)+
𝑉2
𝑍2sin(𝜔𝑡 − 𝜙)
𝐼𝑓2 = −𝑉1
𝑍1 + 𝑍2sin(𝛼 + 𝛿 − 𝜙) +
𝑉2
𝑍1 + 𝑍2sin(𝛼 − 𝜙) +
𝑉2
𝑍2sin(𝜙 − 𝛼) 𝑒
−𝑅2𝐿2
(𝑡−𝑡0)
+𝑉2
𝑍2sin(𝜔𝑡 − 𝜙)
It can be said that the polarity of the wave front depend on the relative
trajectory of the fault current with respect to the pre-fault current, immediately after
the fault. In other words, if 𝑑
𝑑𝑡 of the fault current at t = t0 is greater than that of the
pre-fault current, then the wave front polarity is positive and vice versa. Therefore,
solving the two equations 𝑑𝐼1
𝑑𝑡 𝑡=𝑡0
=𝑑𝐼𝑓1
𝑑𝑡 𝑡=𝑡0
𝑎𝑛𝑑 −𝑑𝐼1
𝑑𝑡 𝑡=𝑡0
=𝑑𝐼𝑓2
𝑑𝑡 𝑡=𝑡0
for 𝛼, would
yield the two inception angles, where the switching of the polarities of the wave-fronts
occur. It can be proven that the solutions given for 𝛼 by both the equations are the
same and can be expressed as,
𝛼 = tan−1 (−sin(𝛿)
𝑘 + cos(𝛿)) , 𝑊ℎ𝑒𝑟𝑒 𝑘 =
𝑉2. 𝑍1
𝑉1. 𝑍2
From this result, it is proven that the wave-front polarities observed at either
end are identical (when the currents out of the busbars are measures as positive), for
all faults internal to the two current measurements. For an actual transmission line,
the equations become much more complicated, but it is observed that the general
behavior of the wave-front polarities remain unchanged. A more comprehensive proof
can be provided by representing the transmission line as a pi-section.